License Plate Detection is an essential part of smart transportation systems which is always ‎expected to be accurate and efficient. The exponential growth of the number of vehicles has led ‎to many problems for vehicle detection in different fields such as traffic control, parking lot, toll ‎highways and so. Using an automatic system for plate number detection of the vehicle, a large ‎number of problems can be solved. One of the best methods for overcoming this problem is using ‎a convolutional neural network approach. They have shown impressive performance in various ‎computer vision tasks. In this work, we tried to introduce a new dataset of Iranian cars' plates. ‎Then using transfer learning and Alexnet, we have proposed a new approach for detection and ‎classifying the plate numbers. We conducted comprehensive experiments on the new dataset. The ‎experimental results show that the proposed method has achieved to 97.35% accuracy.‎

Independent object test is one of the important steps in the object-oriented software test. This kind of test is faced with two problems: firstly, the object which is under test may call methods of other objects and therefore, the independent test of the object becomes impossible. Secondly, the called methods may be time-consuming and as a result the test of the object takes a long time. In order to overcome these problems, a useful method is to use the faked object which simulates the called methods. Faked objects are usually implemented using a table. This table-based implementation results in different problems such as time-consuming table search operation, and more importantly, inability to exact simulation of called methods. Besides, test samples are rare and therefore automatic generation of test samples which span all the code paths within a method has become a challenging problem. In this paper, a new artificial neural network-based faked object is proposed which solves the two above-mentioned problems. This paper contains two proposed sections: in the first section, the operation of linear functions which are used in programs is simulated. In the second section, the best set of input parameters which are needed to train the artificial neural network of faked object is determined optimally using genetic algorithm. The superiority of the proposed methods is confirmed using different experiments for mathematical, logical and discrete functions.

The purpose of this study is to analyze oil and tax revenues in the Iranian economy. Applying the stochastic dynamic general equilibrium approach, a model for the oil exporting economy was simulated. In order to recover the tax system, oil and tax revenues added to the model as a part of government’s revenue. The present study applies Dynare software with Matlab to estimate the model’s parameters by Bayesian statistical techniques based on Monte Carlo method with Markov chain in the form of Metropolis-Hastings algorithm. The period of study is from 1368 to 1396 in a quarterly setting. In order to analyze the shocks, two scenarios were designed. In the first scenario, it was assumed that the government has oil revenues and all oil revenues are spent by the government so that the government does not rely on tax revenues. In the second scenario, it is assumed that 40% of the government's oil revenues are injected into the Development Fund and a percentage of it, is allocated to facilitate the production sector, and the government mostly relies on the taxes. The results show that the tax and oil shock by reducing dependence on oil and reliance on tax revenues, has negative impact on macroeconomic variables in the short term, but in the long run, with increasing tax revenues, production and, consequently, investment, consumption, employment has increased. JEL classification: H21 C13، E17، G5

استراتژی دریایی یکی از مهمترین ابعاد استراتژی ملی قدرت های بزرگ به حساب می آید. این مساله اگرچه با تحول ماهوی قدرت در سیاست جهانی، اهمیتی را که آلفرد ماهان در «تاثیر و نفوذ قدرت دریایی در تاریخ» مطرح کرده بود را از دست داده است اما در دو دهه اخیر مجددا به عرصه رقابت ژئوپلیتیک قدرت های بزرگ بازگشته است. در این میان، اقیانوس هند و شکل گیری ائتلاف بندی های جدیدی مانند کواد و آکوس در آن را می توان نقطه عزیمت جدیدی در بازی قدرت های جهانی دانست. در این راستا، این مقاله به این سوال اصلی پرداخته است که اقیانوس هند چه جایگاهی در معادلات ژئوپلیتیک قدرت های دریایی با تاکید بر گروه بندی های کواد و آکوس دارد؟ یافته های این پژوهش که به روش توصیفی- تحلیلی صورت گرفته نشان دهنده انتقال تدریجی مرکز ثقل اقتصاد و سیاست جهانی از اقیانوس اطلس به اقیانوس هند-آرام است که در پرتو ائتلاف بندی های نظامی و امنیتی جدید به تشدید رقابت و تنش میان قدرت های بزرگ دریایی منجر خواهد شد.

In the field of cyber defense, researchers always suffer from the lack of a proper dataset to evaluate their proposed theories and methods. Unfortunately, in the various datasets existing in cyber defense scope, the ground truth is ambiguous, and the scenarios used by the attackers to carry out the attacks are unclear. This will lead to a serious challenge to the verification of methods and researches in this area. In this paper, a method is proposed by which a new database can be generated with the explisit ground truth and predetermined scenarios for multistage cyber attacks. In this method, a cyber attack guidance template is used to determine the various stages of the attacks and an attack scenario generator is also used to determine the scenarios used by simulated attackers. Network topology is considered as input, and random variables are used to create variety in simulator performances. Also, in the proposed method, various techniques such as fuzzy c-means for clustering, and artificial neural networks for classification are used. To set the simulator parameters, the CDX dataset is used and its ability to create a new dataset of multistage cyber attacks is well illustrated. To evaluate the proposed method, scoring by SME's is used, and finally with the mean score of 90. 7, it had been approved.

Persian romantic verses as texts composed over the centuries have common characteristics which upon their structural analysis one can analyze their inter-textual connections. Attempt is made in this article to describe the farewell scenes of lovers in order to arrive at their common converse in these texts, since the main theme is love and this fundamental feature confers a mutual function to the lovers’ death. According to the inter textual theory, the texts with common theme in different historical periods are interconnected and their images, descriptions and their poetic thoughts are emanated from a common language even if at first they seem distinct. This common language provides the logic and the basis of the colloquy and living in such circumstances. In other words the language of love necessitates that farewell scenes have common language and ambiance although differences arise in accordance with the story and the relationship between the lover and the beloved. In one place, after the death of his beloved, the lover stands by her grave and weeps till his own death. And in other place, he opens his chest with a dagger. It is such that in studying the farewell scenes one arrives at the subject of synchronic death of the lovers. The focal point in these verses is the bitterness and the dilemma of separation which the lovers cannot tolerate and shortly after the other say farewell to life and end the dialogue of love.

Speech databases are part of the concatenative text-to-speech (TTS) systems. Quality of the databases and the way of their development play a significant role in the naturalness of the synthesized speech. This paper introduces a syllable speech database and a diphone speech database and describing their characteristics. The phases in their development are explained. Then differences and advantages of these two databases are discussed using Persian text-to-speech synthesis system “Gooya”.

This paper uses a new meta-heuristic called Grey Wolf Optimizer (GWO) for classifying sonar data set. The GWO algorithm imitates the leadership hierarchy and hunting mechanism of grey wolves in nature. It also employs four types of grey wolves including alpha, beta, delta and omega for simulating the leadership hierarchy. In addition, the three main steps of hunting including searching for prey, encircling prey and attacking prey, are simulated. The algorithm is then benchmarked on 23 well-known test functions and the results are compared with Particle Swarm Optimization (PSO). The results show that the GWO algorithm provides better results in finding total minimum of functions, convergence speed and local minima avoidance compared to PSO. In addition, in this paper a real application of proposed method in the field of sonar data set classification is presented. The results show that the designed classifier inspired by grey wolves can classify the sonar data with accuracy of 96.67%; whereas the PSO presents the accuracy of 92.23%.

Extensive research has been conducted in the field of music signal processing which targeted context-based music information retrieval. Unfortunately, research on the computer-based processing of the traditional Persian music is rare, which is due to lack of standard databases. In this paper, a database, named Nava, is introduced for two basic tasks of the traditional Persian music field, Dastgah classification and instrument recognition. In terms of instrument, Dastgah and artist, Nava has enough comprehensiveness and variety. It contains the sound of five common traditional instruments played by 40 artists in seven Dastgahs. In order to address the two mentioned basic tasks, a system is proposed which extracts a sequence of Mel frequency cepstral coefficients (MFCC) feature vectors from input music signal and then converts it to a fixed-length feature vector using i-vector technique. In the classification stage, the extracted i-vector is fed into a support vector machine classifier. The best obtained accuracy on the Nava database for the Dastgah classification and instrument recognition are about 34% and 98% respectively, which indicates the difficulty of the former in comparison with the latter.

In the past two decades, the applications of computational neuroscience have been increasingly growing. Breaking the neural code is a crucial open problem in computational neuroscience. Various research groups attempt to provide an efficient method to decode human brain activity using fMRI data. The output of these methods is a computational model that can assign brain signals to an external stimulus; in this study, visual object recognition has been investigated. The brain decoders are used in many applications, such as the brain-computer interface or detecting specific mental illnesses. In general, brain fMRI data have a high spatial and temporal resolution that increases the number of features of the problem. Proper feature extraction from brain images is a challenging and time-consuming process. Consequently, the convergence of learning algorithms takes a long time to create an appropriate model. So, breaking down the feature space is highly recommended. We proposed new multi-view learning to solve the brain decoding problem. This approach splits the feature space based on mutual information and finds an appropriate ensemble classification model that detects the related visual object to neural activities in the brain. The proposed method clusters the feature space based on mutual information and splits it into coherent sub-spaces, views. For each feature view, a support vector machine model is learned in parallel; the used SVM version can generate a vector of probabilities for each class. At the test phase, the feature space of test data is divided similarly to the training data, and each model generates a probabilistic vector for the test instances. Then, these vectors are combined in the decision profile matrix. The decision fusion is employed by the ordered weighted averaging (OWA) approach. The proposed multi-view learning methods achieved higher accuracy rates than the single view model. The main advantage of the MV model is that it can run in parallel, making it counterproductive to deal with the high-dimensional problems based on the divide and conquer strategy. The optimization phase to detect the most acceptable parameters for each model is obtained using the simulated annealing, SA, algorithm. We have employed three real fMRI datasets of the human brain to assess the proposed method, obtained from the Openneuro website. Also, the leave-one-run-out cross-validation approach has been carried out to evaluate the proposed method in the intra-subject scenario. Criteria such as accuracy rate and confusion matrix have been undertaken to analyze the results. The single feature view obtains an accuracy rate of more than 50%. While in the ensemble model, the accuracy rate in most subjects is more than 90%.