Sharif Digital Repository

Artificial Life can be used as an agent training approach in large state spaces. This paper presents an artificial life method to increase the training speed of some speech recognizer agents which where previously trained by genetic algorithms. Using this approach, vertical training (genetic mutations and selection) is combined with horizontal training (individual learning through reinforcement learning) and results in a much faster evolution than simple genetic algorithm. The approach is tested and a comparison with GA cases on a standard speech data base is presented. COPYRIGHT © ENFORMATIKA  

In this paper a novel method called Extended Two-Dimensional PCA (E2DPCA) is proposed which is an extension to the original 2DPCA. We state that the covariance matrix of 2DPCA is equivalent to the average of the main diagonal of the covariance matrix of PCA. This implies that 2DPCA eliminates some covariance information that can be useful for recognition. E2DPCA instead of just using the main diagonal considers a radius of r diagonals around it and expands the averaging so as to include the covariance information within those diagonals. The parameter r unifies PCA and 2DPCA. r=1 produces the covariance of 2DPCA, r=n that of PCA. Hence, by controlling r it is possible to control the... 

The joint analysis of multiple data sets to extract their interdependency information has wide applications in biomedical and health informatics. In this paper, we propose an algorithm to extract joint and individual sources of multisubject data sets by using a deflation-based procedure, which is referred to as joint/individual thin independent component analysis (JI-ThICA). The proposed algorithm is based on two cost functions utilizing higher order cumulants to extract joint and individual sources. Joint sources are discriminated by fusing signals of all subjects, whereas individual sources are extracted separately for each subject. Furthermore, JI-ThICA algorithm estimates the number of... 

Boosting is a general method for improving the accuracy of any given learning algorithm. In this paper, we have proposed the boosting method for face recognition (FR) that improves the linear discriminant analysis (LDA)-based technique. The improvement is achieved by incorporating the regularized LDA (R-LDA) technique into the boosting framework. R-LDA is based on a new regularized Fisher's discriminant criterion, which is particularly robust against the small sample size problem compared to the traditional one used in LDA. The AdaBoost technique is utilized within this framework to generalize a set of simple FR subproblems and their corresponding LDA solutions and combines the results from... 

In this paper a new solution for the single sample problem in low resolution face recognition is proposed. The proposed solution uses an enhanced virtual pose generation method to extend the number of face images of each identity. Using a top-right face image of an identity in the gallery, the method generates other poses of the same identity. Face images are represented as a set of local patches. In order to avoid image alignment problems, patches in the first and second pose are clustered. For each cluster the mapping between the patches of the two poses is learned. Experimental results show superior subjective and objective performance of the proposed method on CASPEAL database compared... 

In this study knee joint vibroarthrographic (VAG) signals are recorded during active knee movements, which are essentially non-stationary. Because of this nature, common frequency methods are unable to represent the signals, accurately. Both time-frequency and time-scale transforms are used in this research which are good tools for studying non-stationary signals. By optimizing the utilized transforms, it was concluded that the wavelet packet, having the ability of multiresolutional analysis, is a more promising method to extract features from the VAG signals. The performance of different feature extraction techniques were compared by using three new recorded and extensive databases,... 

This paper presents a method to determine material of an unknown sample object. The main objective of this study is to design a database for specifying material of an object. We produce the database for different materials which is subjected to different forces. For this purpose we use a Polyvinidilene Fluoride (PVDF) sensor which is a piezoelectric material. Also we study the effect of changing place of sensor on our study. The detailed design was performed using finite element method analysis. Furthermore, if we have an object which we do not know its material by use of this database we can find out what this object is and how much its Yanoung's modules is. This study will be suitable for... 

In recent years, end-to-end approaches to automatic speech recognition have received considerable attention as they are much faster in terms of preparing resources. However, conventional multistage approaches, which rely on a pipeline of training hidden Markov models (HMM)-GMM models and tree-building steps still give the state-of-the-art results on most databases. In this study, we investigate flat-start one-stage training of neural networks using lattice-free maximum mutual information (LF-MMI) objective function with HMM for large vocabulary continuous speech recognition. We thoroughly look into different issues that arise in such a setup and propose a standalone system, which achieves... 

In cloud computing, services play key roles. Services are well defined and autonomous components. Nowadays, the demand of using Fuzzy inference as a service is increasing in the domain of complex and critical systems. In such systems, along with the development of the software, the cost of detecting and fixing software defects increases. Therefore, using formal methods, which provide clear, concise, and mathematical interpretation of the system, is crucial for the design of these Fuzzy systems. To obtain this goal, we introduce the Fuzzy Inference Cloud Service (FICS) and propose a novel discipline for formal modeling of the FICS. The FICS provides the service of Fuzzy inference to the... 

In the present study, a prediction model was derived for the effective angle of shearing resistance (φ′) of soils using a novel hybrid method coupling genetic programming (GP) and orthogonal least squares algorithm (OLS). The proposed nonlinear model relates φ′ to the basic soil physical properties. A comprehensive experimental database of consolidated-drained triaxial tests was used to develop the model. Traditional GP and least square regression analyses were performed to benchmark the GP/OLS model against classical approaches. Validity of the model was verified using a part of laboratory data that were not involved in the calibration process. The statistical measures of correlation... 

Dynamic thermal rating (DTR) of transmission system facilities is a way to maximally realize the equipment capacities while not threatening their health. With regards to transmission lines, the allowable current of conductors is forecasted based on the environmental situations expected in some forthcoming time periods. Due to the fact that weather conditions continuously vary, sampling points are very limited against many line spans, and the measurements have an inherent error, uncertainties must be appropriately included in the DTR determination. This paper adopts the fuzzy theory as a strong and simple tool to model uncertainties in the DTR calculation. Since DTR intends to determine the... 

A new face recognition method is proposed in this paper. The proposed method is based on fuzzy regularized linear discriminant analysis (FR-LDA) and combines the regularized linear discriminant analysis (R-LDA) and the fuzzy set theory. R-LDA is based on a new regularized Fisher's discriminant criterion, which is particularly robust against the small sample size problem compared to the traditional one used in LDA. In the proposed method, we calculate the membership degree matrix by Fuzzy K-nearest neighbor (FKNN) and then incorporate the membership degree into the definition of the between-class and within-class scatter matrices and get the fuzzy between-class and within-class scatter... 

Gelatin-based functional films were prepared by the addition of grapefruit seed extract (GSE, 5 wt% based on gelatin) and various amounts of TiO2 (0.5, 1.0, 3.0, and 5.0 wt% based on gelatin). TiO2 was evenly dispersed in the gelatin film, but the film surface roughness was increased as the concentration of TiO2 increased. The mechanical strength and water contact angle (WCA) of the composite film were the highest, while the water vapor permeability (WVP) was the lowest when 0.5 wt% TiO2 was used. The addition of GSE slightly reduced the UV light transmittance, but the addition of TiO2 almost completely prevented the UV light transmission. The addition of GSE and TiO2 did not significantly... 

Automatic translation of a high-level specification language to an executable implementation would be highly useful in maximizing the benefits of formal methods. We will introduce a set of translation functions to fill the specification-implementation gap in the domain of database applications. Because the mathematical foundation of Z has many properties in common with SQL, a direct mapping from Z to SQL structures can be found. We derive a set of translation functions from Z to SQL for the generation of a database. The proposed methodology results in reducing the expenses and duration of the software development, and also, prevents the errors originated from the manual translation of... 

Aims: Due to large sized genetic database of population in a given society, researchers face serious problem to analyze genetic data and provide accurate genetic identification. Variation in molecular markers such as SNPs, mtDNAs, STRs and Y-chromosome are utilized for different purposes, including genetic identification. In our country due to natural disasters and imposed war, the use of this technology was considered and according to the requirements, an optimal database was designed that has the capability of analyzing genetic data. Methods: After obtaining individual genetic information, a software was designed for the analysis of genetic information as well as to serve as a common... 

Graph Matching (GM) plays an essential role in computer vision and machine learning. The ability of using pairwise agreement in GM makes it a powerful approach in feature matching. In this paper, a new formulation is proposed which is more robust when it faces with outlier points. We add weights to the one-to-one constraints, and modify them in the process of optimization in order to diminish the effect of outlier points in the matching procedure. We execute our proposed method on different real and synthetic databases to show both robustness and accuracy in contrast to several conventional GM methods  

In this paper, we propose a mapping from low level feature space to the semantic space drawn by the users through relevance feedback to enhance the performance of current content based image retrieval (CBIR) systems. The proposed approach makes a rule base for its inference and configures it using the feedbacks gathered from users during the life cycle of the system. Each rule makes a hypercube (HC) in the feature space corresponding to a semantic concept in the semantic space. Both short and long term strategies are taken to improve the accuracy of the system in response to each feedback of the user and gradually bridge the semantic gap. A scoring paradigm is designed to determine the... 

There is a growing attention in subspace learning using tensor-based approaches in high dimensional spaces. In this paper we first indicate that these methods suffer from the Heteroscedastic problem and then propose a new approach called Heteroscedastic Multilinear Discriminant Analysis (HMDA). Our method can solve this problem by utilizing the pairwise chernoff distance between every pair of clusters with the same index in different classes. We also show that our method is a general form of Multilinear Discriminant Analysis (MDA) approach. Experimental results on CMU-PIE, AR and AT&T face databases demonstrate that the proposed method always perform better than MDA in term of classification... 

Electrocardiogram is one of the most prominent cardiac signals being capable to be utilized for medical uses such as arrhythmia detection. Over the years, the feasibility of using this signal for human identification issue has been investigated, and some methods have been proposed. In this Paper a novel approach is proposed for electrocardiogram (ECG) based human identification using Extended Kalman Filter (EKF). The innovation signal of EKF has been considered as feature which is used to classify different subjects. In this paper a general issue, human identification, is summarized to a classification problem in which the proposed features of each subject is calculated, and the... 

We propose a novel image steganalysis method, based on singular value decomposition and noise estimation, for the spatial domain LSB embedding families. We first define a content independence parameter, DS, that is calculated for each LSB embedding rate. Next, we estimate the DS curve and use noise estimation to improve the curve approximation accuracy. It is shown that the proposed approach gives an estimate of the LSB embedding rate, as well as information about the existence of the embedded message (if any). The proposed method can effectively be applied to a wide range of the image LSB steganography families in spatial domain. To evaluate the proposed scheme, we applied the method to a...