Abstract: This paper proposes a neural network weights and
topology optimization using genetic evolution and the
backpropagation training algorithm. The proposed crossover and
mutation operators aims to adapt the networks architectures and
weights during the evolution process. Through a specific inheritance
procedure, the weights are transmitted from the parents to their
offsprings, which allows re-exploitation of the already trained
networks and hence the acceleration of the global convergence of the
algorithm. In the preprocessing phase, a new feature extraction
method is proposed based on Legendre moments with the Maximum
entropy principle MEP as a selection criterion. This allows a global
search space reduction in the design of the networks. The proposed
method has been applied and tested on the well known MNIST
database of handwritten digits.
Abstract: To produce sugar and ethanol, sugarcane processing
generates several agricultural residues, being straw and bagasse is
considered as the main among them. And what to do with this
residues has been subject of many studies and experiences in an
industry that, in recent years, highlighted by the ability to transform
waste into valuable products such as electric power. Cellulose is the
main component of these materials. It is the most common organic
polymer and represents about 1.5 x 1012 tons of total production of
biomass per year and is considered an almost inexhaustible source of
raw material. Pretreatment with mineral acids is one of the most
widely used as stage of cellulose extraction from lignocellulosic
materials for solubilizing most of the hemicellulose content. This
study had as goal to find the best reaction time of sugarcane bagasse
pretreatment with sulfuric acid in order to minimize the losses of
cellulose concomitantly with the highest possible removal of
hemicellulose and lignin. It was found that the best time for this
reaction was 40 minutes, in which it was reached a loss of
hemicelluloses around 70% and lignin and cellulose, around 15%.
Over this time, it was verified that the cellulose loss increased and
there was no loss of lignin and hemicellulose.
Abstract: Persian (Farsi) script is totally cursive and each character is written in several different forms depending on its former and later characters in the word. These complexities make automatic handwriting recognition of Persian a very hard problem and there are few contributions trying to work it out. This paper presents a novel practical approach to online recognition of Persian handwriting which is based on representation of inputs and patterns with very simple visual features and comparison of these simple terms. This recognition approach is tested over a set of Persian words and the results have been quite acceptable when the possible words where unknown and they were almost all correct in cases that the words where chosen from a prespecified list.
Abstract: In this paper, a new face recognition method based on
PCA (principal Component Analysis), LDA (Linear Discriminant
Analysis) and neural networks is proposed. This method consists of
four steps: i) Preprocessing, ii) Dimension reduction using PCA, iii)
feature extraction using LDA and iv) classification using neural
network. Combination of PCA and LDA is used for improving the
capability of LDA when a few samples of images are available and
neural classifier is used to reduce number misclassification caused by
not-linearly separable classes. The proposed method was tested on
Yale face database. Experimental results on this database
demonstrated the effectiveness of the proposed method for face
recognition with less misclassification in comparison with previous
methods.
Abstract: Among various testing methodologies, Built-in Self-
Test (BIST) is recognized as a low cost, effective paradigm. Also,
full adders are one of the basic building blocks of most arithmetic
circuits in all processing units. In this paper, an optimized testable 2-
bit full adder as a test building block is proposed. Then, a BIST
procedure is introduced to scale up the building block and to generate
a self testable n-bit full adders. The target design can achieve 100%
fault coverage using insignificant amount of hardware redundancy.
Moreover, Overall test time is reduced by utilizing polymorphic
gates and also by testing full adder building blocks in parallel.
Abstract: The amount and heterogeneity of data in biomedical research, notably in interdisciplinary research, requires new methods for the collection, presentation and analysis of information. Important data from laboratory experiments as well as patient trials are available but come out of distributed resources. The Charite Medical School in Berlin has established together with the German Research Foundation (DFG) a new information service center for kidney diseases and transplantation (Open European Nephrology Science Centre - OpEN.SC). The system is based on a service-oriented architecture (SOA) with main and auxiliary modules arranged in four layers. To improve the reuse and efficient arrangement of the services the functionalities are described as business processes using the standardised Business Process Execution Language (BPEL).
Abstract: There has been a growing interest in implementing humanoid avatars in networked virtual environment. However, most existing avatar communication systems do not take avatars- social backgrounds into consideration. This paper proposes a novel humanoid avatar animation system to represent personalities and facial emotions of avatars based on culture, profession, mood, age, taste, and so forth. We extract semantic keywords from the input text through natural language processing, and then the animations of personalized avatars are retrieved and displayed according to the order of the keywords. Our primary work is focused on giving avatars runtime instruction from multiple natural languages. Experiments with Chinese, Japanese and English input based on the prototype show that interactive avatar animations can be displayed in real time and be made available online. This system provides a more natural and interesting means of human communication, and therefore is expected to be used for cross-cultural communication, multiuser online games, and other entertainment applications.
Abstract: Question answering (QA) aims at retrieving precise information from a large collection of documents. Most of the Question Answering systems composed of three main modules: question processing, document processing and answer processing. Question processing module plays an important role in QA systems to reformulate questions. Moreover answer processing module is an emerging topic in QA systems, where these systems are often required to rank and validate candidate answers. These techniques aiming at finding short and precise answers are often based on the semantic relations and co-occurrence keywords. This paper discussed about a new model for question answering which improved two main modules, question processing and answer processing which both affect on the evaluation of the system operations. There are two important components which are the bases of the question processing. First component is question classification that specifies types of question and answer. Second one is reformulation which converts the user's question into an understandable question by QA system in a specific domain. The objective of an Answer Validation task is thus to judge the correctness of an answer returned by a QA system, according to the text snippet given to support it. For validating answers we apply candidate answer filtering, candidate answer ranking and also it has a final validation section by user voting. Also this paper described new architecture of question and answer processing modules with modeling, implementing and evaluating the system. The system differs from most question answering systems in its answer validation model. This module makes it more suitable to find exact answer. Results show that, from total 50 asked questions, evaluation of the model, show 92% improving the decision of the system.
Abstract: Detection of player identity is challenging task in sport video content analysis. In case of soccer video player number recognition is effective and precise solution. Jersey numbers can be considered as scene text and difficulties in localization and recognition appear due to variations in orientation, size, illumination, motion etc. This paper proposed new method for player number localization and recognition. By observing hue, saturation and value for 50 different jersey examples we noticed that most often combination of low and high saturated pixels is used to separate number and jersey region. Image segmentation method based on this observation is introduced. Then, novel method for player number localization based on internal contours is proposed. False number candidates are filtered using area and aspect ratio. Before OCR processing extracted numbers are enhanced using image smoothing and rotation normalization.
Abstract: Oil debris signal generated from the inductive oil
debris monitor (ODM) is useful information for machine condition
monitoring but is often spoiled by background noise. To improve the
reliability in machine condition monitoring, the high-fidelity signal
has to be recovered from the noisy raw data. Considering that the noise
components with large amplitude often have higher frequency than
that of the oil debris signal, the integral transform is proposed to
enhance the detectability of the oil debris signal. To cancel out the
baseline wander resulting from the integral transform, the empirical
mode decomposition (EMD) method is employed to identify the trend
components. An optimal reconstruction strategy including both
de-trending and de-noising is presented to detect the oil debris signal
with less distortion. The proposed approach is applied to detect the oil
debris signal in the raw data collected from an experimental setup. The
result demonstrates that this approach is able to detect the weak oil
debris signal with acceptable distortion from noisy raw data.
Abstract: DNA microarray technology is widely used by
geneticists to diagnose or treat diseases through gene expression.
This technology is based on the hybridization of a tissue-s DNA
sequence into a substrate and the further analysis of the image
formed by the thousands of genes in the DNA as green, red or yellow
spots. The process of DNA microarray image analysis involves
finding the location of the spots and the quantification of the
expression level of these. In this paper, a tool to perform DNA
microarray image analysis is presented, including a spot addressing
method based on the image projections, the spot segmentation
through contour based segmentation and the extraction of relevant
information due to gene expression.
Abstract: IPsec protocol[1] is a set of security extensions
developed by the IETF and it provides privacy and authentication
services at the IP layer by using modern cryptography. In this paper,
we describe both of H/W and S/W architectures of our router system,
SRS-10. The system is designed to support high performance routing
and IPsec VPN. Especially, we used Cavium-s CN2560 processor to
implement IPsec processing in inline-mode.
Abstract: The purpose of this paper is to present a Dynamic
Time Warping technique which reduces significantly the data
processing time and memory size of multi-dimensional time series
sampled by the biometric smart pen device BiSP. The acquisition
device is a novel ballpoint pen equipped with a diversity of sensors
for monitoring the kinematics and dynamics of handwriting
movement. The DTW algorithm has been applied for time series
analysis of five different sensor channels providing pressure,
acceleration and tilt data of the pen generated during handwriting on
a paper pad. But the standard DTW has processing time and memory
space problems which limit its practical use for online handwriting
recognition. To face with this problem the DTW has been applied to
the sum of the five sensor signals after an adequate down-sampling
of the data. Preliminary results have shown that processing time and
memory size could significantly be reduced without deterioration of
performance in single character and word recognition. Further
excellent accuracy in recognition was achieved which is mainly due
to the reduced dynamic time warping RDTW technique and a novel
pen device BiSP.
Abstract: Biclustering is a very useful data mining technique for
identifying patterns where different genes are co-related based on a
subset of conditions in gene expression analysis. Association rules
mining is an efficient approach to achieve biclustering as in
BIMODULE algorithm but it is sensitive to the value given to its
input parameters and the discretization procedure used in the
preprocessing step, also when noise is present, classical association
rules miners discover multiple small fragments of the true bicluster,
but miss the true bicluster itself. This paper formally presents a
generalized noise tolerant bicluster model, termed as μBicluster. An
iterative algorithm termed as BIDENS based on the proposed model
is introduced that can discover a set of k possibly overlapping
biclusters simultaneously. Our model uses a more flexible method to
partition the dimensions to preserve meaningful and significant
biclusters. The proposed algorithm allows discovering biclusters that
hard to be discovered by BIMODULE. Experimental study on yeast,
human gene expression data and several artificial datasets shows that
our algorithm offers substantial improvements over several
previously proposed biclustering algorithms.
Abstract: Research has suggested that implicit learning tasks
may rely on episodic processing to generate above chance
performance on the standard classification tasks. The current
research examines the invariant features task (McGeorge and Burton,
1990) and argues that such episodic processing is indeed important.
The results of the experiment suggest that both rejection and
similarity strategies are used by participants in this task to
simultaneously reject unfamiliar items and to accept (falsely) familiar
items. Primarily these decisions are based on the presence of low or
high frequency goal based features of the stimuli presented in the
incidental learning phase. It is proposed that a goal based analysis of
the incidental learning task provides a simple step in understanding
which features of the episodic processing are most important for
explaining the match between incidental, implicit learning and test
performance.
Abstract: The textile industry produces highly coloured
effluents containing polar and non-polar compounds. The textile mill
run by the Assam Polyester Co-operative Society Limited (APOL) is
situated at Rangia, about 55 km from Guwahati (26011' N, 91047' E)
in the northern bank of the river Brahmaputra, Assam (India). This
unit was commissioned in June 1988 and started commercial
production in November 1988. The installed capacity of the weaving
unit was 8000 m/day and that of the processing unit was 20,000
m/day. The mill has its own dyeing unit with a capacity of 1500-2000
kg/day. The western side of the mill consists of vast agricultural land
and the far northern and southern side of the mill has scattered human
population. The eastern side of the mill has a major road for
thoroughfare. The mill releases its effluents into the agricultural land
in the western side of the mill. The present study was undertaken to
assess the impact of the textile mill on surface soil quality in and
around the mill with particular reference to Cr, Mn, Ni and Zn.
Surface soil samples, collected along different directions at 200, 500
and 1000 m were digested and the metals were estimated with
Atomic Absorption Spectrophotometer. The metals were found in the
range of: Cr 50.9 – 105.0 mg kg-1, Mn 19.2- 78.6 mg kg-1, Ni 41.9 –
50.6 mg kg-1 and Zn 187.8 – 1095.8 mg kg-1. The study reveals
enrichment of Cr, Mn, Ni and Zn in the soil near the textile mill.
Abstract: In this paper we present an off line system for the
recognition of the handwritten numeric chains. Our work is divided
in two big parts. The first part is the realization of a recognition
system of the isolated handwritten digits. In this case the study is
based mainly on the evaluation of neural network performances,
trained with the gradient back propagation algorithm. The used
parameters to form the input vector of the neural network are
extracted on the binary images of the digits by several methods: the
distribution sequence, the Barr features and the centred moments of
the different projections and profiles. The second part is the
extension of our system for the reading of the handwritten numeric
chains constituted of a variable number of digits. The vertical
projection is used to segment the numeric chain at isolated digits and
every digit (or segment) will be presented separately to the entry of
the system achieved in the first part (recognition system of the
isolated handwritten digits). The result of the recognition of the
numeric chain will be displayed at the exit of the global system.
Abstract: ''Cocktail party problem'' is well known as one of the human auditory abilities. We can recognize the specific sound that we want to listen by this ability even if a lot of undesirable sounds or noises are mixed. Blind source separation (BSS) based on independent component analysis (ICA) is one of the methods by which we can separate only a special signal from their mixed signals with simple hypothesis. In this paper, we propose an online approach for blind source separation using the sliding DFT and the time domain independent component analysis. The proposed method can reduce calculation complexity in comparison with conventional methods, and can be applied to parallel processing by using digital signal processors (DSPs) and so on. We evaluate this method and show its availability.
Abstract: Clustering is the process of subdividing an input data set into a desired number of subgroups so that members of the same subgroup are similar and members of different subgroups have diverse properties. Many heuristic algorithms have been applied to the clustering problem, which is known to be NP Hard. Genetic algorithms have been used in a wide variety of fields to perform clustering, however, the technique normally has a long running time in terms of input set size. This paper proposes an efficient genetic algorithm for clustering on very large data sets, especially on image data sets. The genetic algorithm uses the most time efficient techniques along with preprocessing of the input data set. We test our algorithm on both artificial and real image data sets, both of which are of large size. The experimental results show that our algorithm outperforms the k-means algorithm in terms of running time as well as the quality of the clustering.
Abstract: An approach to develop the FPGA of a flexible key
RSA encryption engine that can be used as a standard device in the
secured communication system is presented. The VHDL modeling of
this RSA encryption engine has the unique characteristics of
supporting multiple key sizes, thus can easily be fit into the systems
that require different levels of security. A simple nested loop addition
and subtraction have been used in order to implement the RSA
operation. This has made the processing time faster and used
comparatively smaller amount of space in the FPGA. The hardware
design is targeted on Altera STRATIX II device and determined that
the flexible key RSA encryption engine can be best suited in the
device named EP2S30F484C3. The RSA encryption implementation
has made use of 13,779 units of logic elements and achieved a clock
frequency of 17.77MHz. It has been verified that this RSA
encryption engine can perform 32-bit, 256-bit and 1024-bit
encryption operation in less than 41.585us, 531.515us and 790.61us
respectively.