Journal of Advances in Computer Research
Volume:4 Issue: 2, Spring 2013

Texture image analysis is one of the most important working realms of image processing in medical sciences and industry. Up to present, different approaches have been proposed for segmentation of texture images. In this paper, we offered unsupervised texture image segmentation based on Markov Random Field (MRF) model. First, we used Gabor filter with different parameters’ (frequency, orientation) values. The output image of this step clarified different textures and then used low pass Gaussian filter for smoothing the image. These two filters were used as preprocessing stage of texture images. In this research, we used K-means algorithm for initial segmentation. In this study, we used Expectation Maximization (EM) algorithm to estimate parameters, too. Finally, the segmentation was done by Iterated Conditional Modes (ICM) algorithm updating the labels and minimizing the energy function. In order to test the segmentation performance, some of the standard images of Brodatz database are used. The experimental results show the effectiveness of the proposed method.

Due to the quick growth of the World Wide Web, retrieval of useful information from the Internet for a particular web user or a group of users becomes very difficult. Recommendation systems using web usage mining help providing an adaptive web environment for the web users. This paper presents a novel approach for page recommendation using fuzzy association rule mining algorithm. This method extracts previous users` access patterns and then employs them to recommend appropriate web pages for the active user. An illustrative example explains this method in details.

Among abundant image denoising methods proposed so far, the use of patch based algorithms have attracted a lot of attention from image processing community. Although these methods are very powerful in presentation of high quality results, the impact of human visual system (HVS) is ignored in sole of them. In this paper the human visual geometry is used in preparation of a new method for image denoising. Several image quality assessment (IQA) criteria, based on HVS, are used to confirm superiority of the proposed method in comparison with other state-of-the-art methods. In addition to denoising quality, the proposed method is fast as a result of dimensionality reduction.

The Gravitational Search Algorithm (GSA) is a novel optimization method based on the law of gravity and mass interactions. It has good ability to search for the global optimum, but its searching speed is really slow in the last iterations. So the hybridization of Particle Swarm Optimization (PSO) and GSA can resolve the aforementioned problem. In this paper, a modified PSO, which the movement of particles is also based on getting away from individual worst solution other than going toward the best ones, is combined with GSA, named (PSOGSA) and is applied on ELD problem. A 6 unit case study considering transmission loss, prohibited zones and ramp rate limits and also a 40 unit system with valve point loading effect has been used to show the feasibility of the method. The results show fast and great convergence compared to the many other previously applied methods.

Acoustic analysis is a proper method in vocal fold pathology diagnosis so that it can complement and in some cases replace the other invasive, based on direct vocal fold observation, methods. There are different approaches and algorithms for vocal fold pathology diagnosis. These algorithms usually have three stages which are Feature Extraction, Feature Reduction and Classification. In this paper initial study of feature extraction and feature reduction in the task of vocal fold pathology diagnosis has been presented. A new type of feature vector, based on wavelet packet decomposition and Mel-Frequency-Cepstral-Coefficients (MFCCs), is proposed. Also a new GA-based method for feature reduction stage is proposed and compared with conventional methods such as Principal Component Analysis (PCA). Gaussian Mixture Model (GMM) is used as a classifier for evaluating the performance of the proposed method. The results show the priority of the proposed method in comparison with current methods.

The large number of applications manages time varying data. Existing database technology seldom supports temporal database, TDB, according to time aspects. These intrinsic temporal database applications rely on such database which stores and retrieves time referenced data. Moreover, applications need to be managed on common data items access simultaneously and to be precluded from inconsistency as soon as possible which is the main task of concurrency controller or CC in short. The method used by CC in typical DB differs from its attitude with TDB. The variety algorithms were proposed regarding to TDB properties by reduction of granule size and decreasing the rate of conflicts to satisfy good performance, but none of them has achieved robust results. There are two categories of CC such as pessimistic and optimistic. In this paper new approach, with considering the TDB aspects, based on optimistic method has been suggested. It reclines the size of granule as data item appropriately and recognizes the conflicts swiftly. Consequently, we compare our proposed algorithm with pervasive 2PL-pessimistic approach. The outcome shows that new proposed algorithm has high degree of trade off with satisfying nearconflict time detection and high rate of parallelism metrics.

The problem of series two-block disjoint decomposition of completely specified Boolean functions is considered. Analysis and investigation of such systems are very important in logical design context. Recently, a good method for solving this problem was suggested which has been based on the ternary matrix cover approach. Using this method a computer program was developed. This paper is focused on decomposability of a system of Boolean functions. The experiments were done on generated systems and standard benchmarks. In decomposable systems, the total number of solutions and the time elapsed to achieve them are inspected. The total number of solutions among all partitions for investigated systems, ranged between 3% and 87% in generated systems and also, 1% and 96% in standard benchmarks.

The problem of automatic signature recognition has received little attention in comparison with the problem of signature verification, despite its potential applications for many business processes and can be used effectively in paperless office projects. This paper presents model-based off-line signature recognition with rotation invariant features. Non-linear rotation of signature patterns is one of the major difficulties to be solved in this problem. The proposed system is designed based on support vector machines (SVM) classifier technique and rotation invariant structure feature to tackle the problem. Our designed system consists of three stages: the first is preprocessing stage, the second is feature extraction stage and the last is SVM classifier stage. Experimental results demonstrated that the proposed methods were effective to improve recognition accuracy.