Accepted papers
- Artificial neural network based forward kinematics solution for planar parallel
manipulators passing through singular configuration
Ammar H. Elsheikh, Ezzat A. Showaib and Abd Elwahed M. Asar, Tanta University, Tanta, Egypt
ABSTRACT
It is well known that, the main drawback of parallel manipulators is the existence of singularities within its workspace,
an Artificial Neural Network (ANN) based solution is proposed in this paper. The proposed approach can certainly learn the
input-output data and discover the non-linear relationships which are inherent in the training data. Additionally, the proposed
approach can provide solution of the forward kinematic problem with reasonable errors at and in the vicinity of kinematic
singularities. The approach is implemented for the 3-RPR, 3-PRR, and 3-RRR planar parallel manipulators.
-
The design of a real time accelerometer based baby sleeping position monitoring and correction system
Sudharsanan.S and Karthikeyan.B, Vit University,India.ABSTRACT
The purpose of this study is to present a novel real- time baby sleeping position monitoring and
correction system to avoid Sudden Infant Death Syndrome (SIDS). The sensing module composed of 3-axis
accelerometer is placed on the baby's forehead to monitor its position during sleeping by calculating the angles
between gravity vector and its axis. System is driven by an inexpensive and low power microcontroller. The
baby's sleeping position information is integrated to the correction system, consists of a Cradle and two servo
motor. Cradle is designed to correct the sleeping position of the baby, and to avoid the SIDS.
- A New Approach for Zone Identification on Printed Gujarati Text: Vertical Bar Method
Shweta Agravat and Mukesh Goswami, Dharmsinh Desai University, Nadiad, India
ABSTRACT
Gujarati language belongs to Indo-Aryan family of languages, widely spo-ken in western Indian state - Gujarat. Gujarati is a multilevel script comprises of 3 differ-ent zones- Upper Zone, Middle Zone and Lower Zone. Several characters in Middle Zone have modifiers attached with it at upper or lower region. To discriminate between charac-ter and modifier, Zone boundaries between all three zones are needed to be identified. Some Devnagari languages like, Hindi and Bangla have existing OCR systems. Other Devnagari languages have existence of Shirorekha. Presence of Shirorekha gives promi-nent pick in horizontal profile that makes Upper and Middle Zone separation easy. But in absence of Shirorekha, Zone boundary Identification methods used in other Devnagari script OCR systems will not be useful for Gujarati Script, and Zone Boundary Identifica-tion becomes a difficult task for Gujarati script. This paper proposes a new Zone Boun-dary Identification Approach based on the Vertical Bar present in Gujarati script. This method is tested on 250 Machine Printed lines and 200 Laser Printed Lines, and accuracy achieved is 80% and 93% respectively. It is expected that this method will reduce the complexity of the Zone Identification methods and provide efficient results.
-
Classification of handwritten gujarati numeral
Archana N. Vyas and Mukesh M. Goswami, Dharmsinh Desai University,Nadiad, India.
ABSTRACT
This paper addresses the problem of recognizing handwritten numerals for Gujarati Language. Three methods are presented for feature extraction. One belongs to the spatial domain and other two belongs to the transform domain. In first technique, a new method has been proposed for spatial domain which is based on Freeman chain code. This method obtains the global direction by considering n x n neighborhood and thus eliminates the noise which occurs due to local direction. In the second method 85 dimensional Fourier descriptors were computed and treated as feature vector and in third feature set DCT coefficients are used as feature vector. These methods are tested with three different classifiers namely KNN, SVM and ANN with back propagation. Various preprocessing steps are applied before classification. Experimental results were evaluated using 10 cross fold validation. The average recognition rate for full data set with modified chain code is 85.67%, 83.63% and 84.89% with KNN, SVM and ANN respectively. The overall recognition rates with DFT are 93.60%, 92.43% and 92.86% with KNN, SVM and ANN respectively where as DCT coefficients provides average recognition rates as 91.03%, 93.00% and 92.07% with KNN, SVM and ANN classifiers respectively.
- Automatic Test Data Generation Using A Genetic Algorithm
Nassima Aleb and Samir Kechid, university of sciences and technologies,algeria
ABSTRACT
The use of metaheuristic search techniques for the automatic generation of test data has been a burgeoning interest for many researchers in recent years. Previous attempts to automate the test generation process have been limited, having been constrained by the size and complexity of software, and the basic fact that in general, test data generation is an undecidable problem. Metaheuristic search techniques offer much promise in regard to these problems. Metaheuristic search techniques are high-level frameworks, which utilize heuristics to seek solutions for combinatorial problems at a reasonable computational cost. In this paper, we present a new evolutionary approach for automated test data generation for structural testing. Our method presents several noteworthy features: It uses a newly defined program modeling allowing an easy program manipulation. Furthermore, instead of affecting a unique value for each input variable, we assign to each input an interval. This representation has the advantage of delimiting first the input value and to refine the interval progressively. In this manner, the search space is explored more efficiently. We use an original fitness function, which expresses truthfully the individual quality. Furthermore, we define a crossover operator allowing to effectively improving individuals.
-
An efficient implementation of genetic algorithms using
product line engineering
Abdelghani Alidra1 and Mohamed Tahar Kimour2, Skikda university1 and Badji Mokhtar-Annaba university2, Algeria.
ABSTRACT
A genetic algorithm (GA) is a metaheuristic
biologically inspired search method that is often used to solve
complex and highly complex problems. When addressing a
specific problem, many parts of the genetic algorithm are
shared whereas many other parts are tailored to the actual
problem. Object oriented mechanisms such as polymorphisme
and heritage are usually used to facilitate the implementation of
genetic algorithms. In the present article we show that this
approach is complex and error prone. We propose a new
approach based on Product Line engineering to facilitate and
optimize the process of developing genetic algorithms solutions.
We believe that our approach will also benefit from recent
advances in SPLE techniques concerning verification,
configuration and dynamic evolution.
-
Off-line Signature Recognition based on Angle
Features and Artificial Neural Network Algorithm
Laila Y. Fannes1 and Dr. Ahmed Y. Ben Sasi2, Libyan Academy1 and College of Industrial Technology2, Misursts, Libya.
ABSTRACT
This research presents a handwritten
signature recognition based on angle feature vector using
Artificial Neural Network (ANN). Each signature image
will be represented by an Angle vector. The feature vector
will constitute the input to the ANN. The collection of
signature images will be divided into two sets. One set will
be used for training the ANN in a supervised fashion. The
other set which is never seen by the ANN will be used for
testing. After training, the ANN will be tested for
recognition of the signature. When the signature is
classified correctly, it is considered correct recognition
otherwise it is a failure. The achieved recognition rate
from this system was 94%.
-
Planning by case-based reasoning based on fuzzy logic
Atmani Baghdad, Benbelkacem Sofia and Benamina Mohamed, University of Oran, Algeria
.
ABSTRACT
The treatment of complex systems often requires the manipulation of vague, imprecise and uncertain information. Indeed, the human being is competent in handling of such systems in a natural way. Instead of thinking in mathematical terms, humans describes the behavior of the system by language proposals. In order to represent this type of information, Zadeh proposed to model the mechanism of human thought by approximate reasoning based on linguistic variables. He introduced the theory of fuzzy sets in 1965, which provides an interface between language and digital worlds. In this paper, we propose a Boolean modeling of the fuzzy reasoning that we baptized Fuzzy-BML and uses the characteristics of induction graph classification. Fuzzy-BML is the process by which the retrieval phase of a CBR is modelled not in the conventional form of mathematical equations, but in the form of a database with membership functions of fuzzy rules.
