SIGNAL AND DATA PROCESSING
Year:2015 | Volume:- | Issue:4 (SERIAL 26)|Papers Count:9

Despite error resilient methods that are applied on video data in transmitter side, occurring error along video data transferring for communication channels is inevitable. Error concealment is a useful method for improving the quality of damaged videos on the receiver side. In this paper, a fast and hybrid boundary matching algorithm is presented for more accurate estimating of damaged motion vectors (MVs) from received video. According to the preference list of error concealment, the proposed algorithm performs the error concealment for each damaged macroblock (MB). In the presented method, the boundary distortion is calculated for each pixel from each candidate MB's boundary with use of proposed hybrid boundary matching criterion. Then, depending on the accuracy of each adjacent boundary from damaged MB, a special weight is given to them through match process. Finally, the list of error concealment preference is updated and the candidate MV with the lowest boundary distortion is selected as the MV of damaged MB. Experimental results show that the proposed algorithm increases the average of PSNR for different test sequences more than 1.8 dB in comparison with reference methods and without significant increasing in calculation time and with improving the quality of reconstructed videos.

The population of elderly people has growing trend in the developed or developing countries. Since the elderly are mainly associated with disability, this group of people exposed to dangerous events such as falling down. It is therefore needed to take care of these people against dangerous events. Intelligent video monitoring is a approach that may be able to give quick notification to the caregivers. For this purpose, in this paper, an approach using a novel tracking method based on modified contour algorithm is presented. The new method is able to conduct tracking and falling down detection in a realistic conditions in the presence of multiple motions. Simulations indicate that the proposed algorithm is able to identify the fall-down event with high accuracy and speed.

GuessAdvantages of HGD attack algorithm over ad-hoc attack is that it is designed algorithmically  and determine attacks are general attacks on stream ciphers. These attacks are classified into ad-hoc and Heuristic Guess and Determine (HGD) attacks. One of the for a large class of stream ciphers while being powerful. In this paper, we use auxiliary polynomials in addition to the original equations as the inputs to the HGD attack on TIPSY and SNOW 1.0 stream ciphers. Based on the concept of guessed basis, the number of guesses in both HGD attack and the improved one on TIPSY is six, however the attack complexity is reduced from O(2102) to O(296). This amount is equal to that of ad-hoc attack, but the size of the guessed basis is improved from seven to six. Also, the complexity of GD attack on SNOW 1.0 of heuristic one with the guessed basis of size 6 and ad-hoc attack with the guessed basis of size 7are O(2202) and O(2224), respectively. However, the complexity and the size of guessed basis of the improved HGD attack are reduced to O(2160) and 5, respectively.

In this paper, we will introduce an intelligent system to edit and spell check Persian texts. The goal is editing and preprocessing Persian texts for natural language processing tasks. This system is based on an expandable and engineering approach and is composed of three subsystems: Persian text editor, spell checker and stemmer. These parts interact with each other to edit texts. To do this, the stemmer subsystem process each word in the text; if the subsystem could not find a stem in the lexicon, the word will be recognized as an incorrect word. Then, the spell checker provides a list of suggestions to correct the wrong word. Subsequently, the editor subsystem edits the text based on the standards of the Academy of Persian Language and Literature. Our evaluation shows nearly 92%, 95% and 96% precision numbers for editor, stemmer and spell checker subsystems, respectively.

In this study, first a supervised version for probabilistic principal component analysis mixture model (SPPCAMM) is proposed. Then, considering projection penalty in learning of a predictive model, a method for face recognition using a dimensionality reduction without loss framework is proposed. In the proposed method, first a locally linear underlying manifold of data samples is obtained using supervised probabilistic principal component analysis mixture model. Then, a support vector machine with projection penalty is trained as the mentioned predictive model using this locally linear underlying manifold. In this way, the benefits of dimensionality reduction are used in the predictive model, while using the projection penalty idea, the loss of useful information is prevented. To train and evaluate the proposed method, well-known face databases are used. Gabor feature extraction method is applied on the face images. The experimental results show that the proposed method has the most average classification accuracy compared to many traditional methods, and also compared to the projection penalty idea used for linear and non-linear kernel-based dimensionality reduction techniques.

Flexibility of woven fabric structure has caused many errors in yarn location detection using customary methods of image processing. On this line, proposing an adaptive method with fabric image properties is concentrated to extract its parameters. In this regards, using meta-heuristic algorithms seems applicable to correspond extraction algorithm of structural parameters to the image conditions. In this study, a new method is proposed for woven fabric image preprocessing and structural texture detection applying compound methods of signal processing, fuzzy clustering and genetic algorithm. Results indicate that proposed method is capable of detecting exact yarn location with mean precision of more than 73 percent in double-layered fabric images with uneven color pattern. In one-layered fabric images with low density weave and invariable color pattern, the mean precision is more than 84 percent.

Gigantic amount of textual data being transferred in web every day. Like other communities, cyberspace is vulnerable to attacks, false information and deception. It becomes increasingly important to design an efficient method to trace identity in this community. In order to investigate the problem of gender identification, we propose 48 features, and design three machine learning algorithms. The results of study showed that AD tree classifier had accuracy up to 73.8%.

In this paper, an automatic method in converting a dependency parse tree into its equivalent phrase structure one is introduced for the Persian language. In the first step, a rule-based algorithm is designed. Then, Persian specific dependency-to-phrase structure conversion rules merge to the algorithm. Subsequently, the Persian dependency treebank with about 30,000 sentences is used as an input for the algorithm and an equivalent phrase structure treebank is extracted. Finally, the statistical Stanford parser is trained using the resulting treebank. Experimental results show a F1 of 96.05% for the conversion algorithm and an F1 of 86.01% for Persian factored model parser.

Word sense disambiguation is the task of identifying the correct sense for the Word in a given context among a finite set of possible senses, and plays an important role in many natural language processing applications such as machine translation, document classification, and information retrieval.In this paper we proposed to use LDA topic model for disambiguating Farsi homograph words with new features. A topic model is a statistical model for extract topics from documents of a corpus. We use Latent Dirichlet Allocation (LDA) that is an unsupervised technique.The system achieved 97% precision for 4 high frequently Farsi homograph words.