Najme Ghanbari

Shuffled Frog Leaping Algorithm and Feature Selection for Improving Recognition Rate of Persian Handwritten Digits Classifier

HOLOS

In this paper, Shuffled Frog Leaping Algorithm is used to improve the recognition rate of Persian... more In this paper, Shuffled Frog Leaping Algorithm is used to improve the recognition rate of Persian handwritten digits. In proposed approach, the effective features in increasing the recognition rate are selected using the Binary Shuffled Frog Leaping Algorithm (BSFLA). By selecting the most suitable features from among all extracted features, the recognition rate is improved and computational costs are also decreased. The fitness function in BSFLA is the number of errors in the Fuzzy classifier which its minimum value is desired. The results indicate that Shuffled Frog Leaping algorithm (SFLA) is more efficient.

Download

A Review of Research Studies on the Recognition of Farsi Alphabetic and Numeric Characters in the Last Decade

Lecture Notes in Electrical Engineering, 2018

The present paper is a review of research on the recognition of Persian alphabetic and numeric ch... more The present paper is a review of research on the recognition of Persian alphabetic and numeric characters in the last ten years. The goal is to provide researchers with the latest research findings for further research after discovering the gaps that still exist in this field, i.e. Optical Character Recognition (OCR) in the Farsi language. However, there has been a dramatic improvement in this field in the last decade as it will be discussed. Although the researchers have achieved very good results, there is still a gap due to the lack of commercial and functional software.

The GSA Algorithm at Two Methods of Feature Selection and Weighting Features to Improve the Recognition Rate of Persian Handwrite Digits with Fuzzy Classifier

International Journal of Computer Applications, 2014

In this paper, using Gravitational search algorithm or GSA recognition rate of Persian handwritte... more In this paper, using Gravitational search algorithm or GSA recognition rate of Persian handwritten digits can be improved. Two methods have been proposed to improve the recognition rate. in the first method, with using of version binary of Gravitational search algorithm or BGSA, we choose optimal features among the overall features extracted. Finding the best feature sets from entire extracted features, not only the number of features and computational burden will be reduced but also recognition rate will be significantly improved. Also In this paper, real version of GSA or RGSA has been used in different way (secondary methods) to improve recognition rate. In this method, instead of choosing some of the features, one random weight has been assigned to each feature. Indeed, feature vector has been multiplied in weight vector to obtain a new feature vector. This Weight vector is obtained with RGSA. After several iterations, RGSA algorithm determines Weight set of features so that Cla...

Download

A Fuzzy Classifier using Bank of Gabor Filters to Facial Images ’ Gender Classification

This paper explores a design-based method to use Gabor filter features and fuzzy system for impro... more This paper explores a design-based method to use Gabor filter features and fuzzy system for improved texture recognition. In this article, we will present an unsupervised approach for facial image gender classification based on texture features.. This system is 1-input, 1-output Mamdani type one, applies facial texture propeties as inputs and reveals the probability of being male face image in output. Choosing proper threshold value, the decision about gender is made. 85.5% obtained classification rate prove the efficiency of proposed method.

A Review of Feature Selection Methods with the Applications in Pattern Recognition in the Last Decade

Lecture Notes in Electrical Engineering, 2018

The present study is a review of recently-done research (in the past 10 years) on the feature sel... more The present study is a review of recently-done research (in the past 10 years) on the feature selection methods and a set of the applications in pattern recognition. The study aimed to introduce the latest research on the feature selection methods and applications. The study findings can be the basis for further and more practical research in this field. Significant advances have been made in the last decade. Particularly in recent years, the evolutionary algorithms related to random methods were widely used to solve feature selection problems.

Particle Swarm Optimization Algorithm-Based Fault Location Using Asynchronous Data Recorded at Both Sides of Transmission Line

1 University of Zabol, Zabol, Electrical Engineering Department, Faculty of Engineering, Iran; E-... more 1 University of Zabol, Zabol, Electrical Engineering Department, Faculty of Engineering, Iran; E-mail: mahdi_ghazy@yahoo.com 2 Shahid Bahonar University of Kerman, Kerman, Faculty of Technical and Engineering, Computer Engineering Department, Iran 3 University of Zabol, Zabol, Faculty of Engineering, Electrical Engineering Department, Iran. PARTICLE SWARM OPTIMIZATION ALGORITHM-BASED FAULT LOCATION USING ASYNCHRONOUS DATA RECORDED AT BOTH SIDES OF TRANSMISSION LINE

Download

Improving Recognition Rate of Persian Handwritten Digits Classifier Using FCM Clustering and Inclined Planes System Optimization Algorithm

In this paper, a method is proposed to increase the recognition rate of the Persian handwritten d... more In this paper, a method is proposed to increase the recognition rate of the Persian handwritten digits classifier. A fuzzy method is used for classification in which there is just one fuzzy rule for each digit [1]. The number of Persian digits is 10, Therefore, the total number of fuzzy rules used in the base method is 10. After implementation, Where the number of fuzzy rules is ten, the recognition rate is 86.21 and 80.19 percent for training and testing samples, respectively. Clustering is used to increase the recognition rate. Training data related to each digit are divided into several clusters and one fuzzy rule is extracted for each cluster. Thus, the number of fuzzy rules is significantly increased (784 rules), since the recognition rate is also increased significantly, increase of fuzzy rules is acceptable. By applying this method, the recognition rate of training and testing samples is increased to 98.97 and 97.15 percent, respectively. FCM method is used for clustering. To...

Download

Optimizing Recognition System of Persian Handwritten Digits

Features selection is one of the issues addressed in pattern recognition. By informed choice of e... more 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 and Binary Genetic Algorithm (BGA), both are population based algorithm series are used to find the best groups of fuzzy recognitions features of handwritten digits. Also In this article, real version of PSO (RPSO) has been used in different way to improve recognition rate. In this method, instead of choosing some of the features, one random weight has been assigned to each feature. Indeed, feature vector has been multiplied in weight vector to obtain a new feature vector. This Weight vector is obtained with RPSO. After several iterations, RPSO algorithm determines Weight set of features so that Classification accuracy increases. Fitness function in these algorithms is the number of fuzzy classifier errors and the aim is to make this value minimum. The obtained results confirmed that the population based algorithm with reducing the number of features and increasing the Rate of recognition have proper performance.

Download

New Steganalysis Method using GLCM and Neural Network

International Journal of Computer Applications, 2012

Steganography is the art of hidden writing and secret communication. The goal of Steganography is... more Steganography is the art of hidden writing and secret communication. The goal of Steganography is to hide a message in a multimedia objet such as image. Steganalysis is the art and science of detecting such the hidden messages. The Gray level Co-occurrence matrix (GLCM) is the matrix containing information about the relationship between values of adjacent pixel in an image. In this paper, we extract features from GLCM that are different between cover image (image without hidden information) and stego image (image with hidden information). In the proposed algorithm, first, we use a combined method of steganography based on both location and conversion to hide the information in the original image and call it image-steg1 image. Then, we hide the information in imagesteg1 again and call it image-steg2. Using GLCM matrix properties, we investigate some different features in the GLCM of the original image and stego images. We can extract features that are different between these images. Features are used for training neural network and the classification step was accomplished using four layers Multi Layer Perceptron (MLP) neural network. We tested our algorithm on 800 standard image databases and we detected 80% of stego images. Therefore, our proposed algorithm efficiency is 80%.

Download

Clustering of Fuzzy Data Based on Particle Swarm Optimization

2022 30th International Conference on Electrical Engineering (ICEE)

Optimizing Recognition System of Persian Handwritten Digits

by seyyed mohammad razavi and Najme Ghanbari

Features selection is one of the issues addressed in pattern recognition. By informed choice of e... more 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 and Binary Genetic Algorithm (BGA), both are population based algorithm series are used to find the best groups of fuzzy recognitions features of handwritten digits. Also In this article, real version of PSO (RPSO) has been used in different way to improve recognition rate. In this method, instead of choosing some of the features, one random weight has been assigned to each feature. Indeed, feature vector has been multiplied in weight vector to obtain a new feature vector. This Weight vector is obtained with RPSO. After several iterations, RPSO algorithm determines Weight set of features so that Classification accuracy increases. Fitness function in these algorithms is the number of fuzzy classifier errors and the aim is to make this value minimum. The obtained results confirmed that the population based algorithm with reducing the number of features and increasing the Rate of recognition have proper performance.

Download

Uploads

Papers by Najme Ghanbari

Log In