Figure 2 - uploaded by Seyyed mohammad Razavi
Content may be subject to copyright.
display of one Chromosomes and How to Select Features
Source publication
Features selection is one of the issues addressed in pattern recognition. By informed choice of effective features on increasing recognition rate among the overall features extracted from, the computational costs can be reduced and deducing unnecessary features are avoided. In this article, the Binary Particle Swarm Optimization (BPSO) algorithm an...
Context in source publication
Context 1
... genetic algorithm is used as one of the efficient and general tool in optimization in most engineering branches. A method is presented to select the best feature set among the entire features set using the genetic algorithm. Genetic search algorithm is a method which initiates the search in parallel and multilaterally from different parts of the problem solving space. This method was introduced in 1973 by John Holland and by publishing the book “Adapting the Natural and Artificial Systems” and then widely used in engineering issues, especially in engineering optimization issues. In genetic algorithm, the principles of Darwin’s natural selection are used to find the optimal formula for predicting or matching the pattern. The genetic search methods are mainly based on the fittest selection process in natural system. By fittest selection, it means the organisms which are more capable and competent to be consistent with nature are more possibly acquired with survival and reproduction and are reached to a higher degree of competence after a few generations. The selection process is conducted by combining the genetic operators including selectivity, reproduction, confluence and mutation. As mentioned before, each database digits of 64 features has been extracted from zoning method. The appropriate features are selected for classifying digits here using genetic algorithm among this 64 features. A binary chromosome length in 64 is defined for these 64 features. If the gene is one, it means using the feature in digits classification and if the gene is zero, it means not using that feature in digits classification (Figure 2). The initial population is randomly created. Then the fitness value is calculated for each chromosome using the fitness function which is the number of errors here. The fitness function is mapped between zero and one. The features of genetic algorithm are given in Table 1. The appropriate fitness function value is obtained by applying genetic algorithm in several generations and selecting a chromosome with the lowest amount. The number of features is reduced from 64 features to 34 features. The initial population is about 20 chromosomes. Table 2 shows the matrix of performance in this ...
Similar publications
The aim of present research is to classify the palmprint images using fuzzy logic for different security aspects. Fuzzy logic is relatively efficient and advanced theory, at present wide use of fuzzy logic is in the classification of remotely sensed images. A branch of biometric, palmprint authentication have increasing attention because palmprint...
Because of the rapid spread of COVID-19 to almost every part of the globe, huge volumes of data and case studies have been made available, providing researchers with a unique opportunity to find trends and make discoveries like never before by leveraging such big data. This data is of many different varieties and can be of different levels of verac...
Plasma jets are widely investigated both in the laboratory and in nature. Astrophysical objects such as black holes, active galactic nuclei, and young stellar objects commonly emit plasma jets in various forms. With the availability of data from plasma jet experiments resembling astrophysical plasma jets, classification of such data would potential...
SARS-CoV-2, like any other virus, continues to mutate as it spreads, according to an evolutionary process. Unlike any other virus, the number of currently available sequences of SARS-CoV-2 in public databases such as GISAID is already several million. This amount of data has the potential to uncover the evolutionary dynamics of a virus like never b...
The study presents an efficient gender classification technique which is based on novel feature selection methods from Haar based wavelet packets and Eigen values of Gabor filters. This form an effective feature vector to be used by the neural network classifier. The method presented in this paper enjoys the features like high speed and low space r...
Citations
... Ghanbari [32] extracted zonal features and then they used fuzzy method for recognition and for better accuracy RPSO algorithm was used for selection the best group for fuzzy method. ...
There are a lot of intensive research on character and numeral recognition for some of popular scripts such as Roman, Arabic, Persian/Farsi, Chinese and Indian. Unfortunately, there is less work on the Urdu language which is used by ¼ population of the world. In this paper we present a review on numeral recognition for Urdu, Arabic and Farsi. Most of the published paper for Urdu language since 2003 and recent publications in the Arabic and Farsi languages have been summarized in this survey paper. The Unicode system and writing glyph of digits have been analyzed in Arabic language and its derivative languages. We also added some future direction of research on numeral recognition system.
... In [10], Particle Swarm Optimization and genetic algorithm have been used for better recognition of Persian handwritten digits. In [11], a hybrid method including neural networks and ant 2 colony optimization has been presented in which neural network has been used as a classifier function used in ant colony optimization (ACO). ...
Curse of dimensionality is one of the biggest challenges in classification problems. High dimensionality of problem increases classification rate and brings about classification error. Selecting an effective subset of features is an important point in analyzing correlation rate in classification issues. The main purpose of this paper is enhancing characters recognition and classification, creating quick and low-cost classes, and eventually recognizing Persian handwritten characters more accurately and faster. In this paper, to reduce feature dimensionality of datasets a hybrid approach using artificial neural network, genetic algorithm and quantum genetic algorithm is proposed that can be used to distinguish Persian handwritten letters. Implementation results show that proposed algorithms are able to reduce number of features by 19% to 49%. They also show that recognition and classification accuracy of resulted subset of features has risen, by 7/31%, comparing to primitive dataset.
