Introduction: The identification of effective factors in air pollution is necessary to plan for decreasing the impacts. Tehran is one of the most polluted cities in Iran and the world, which is constantly affected by air pollution. The traffic zone is attributed to most crowded places in Tehran with some limitations for vehicles. In this study, samples were taken from the limited traffic zone in Tehran. Materials and methods: The sampling was done with high volume sampler for a period of 6-24 h. Preliminary SEM/EDS measurements were performed on some of the samples to get elemental information on individual aerosol particles. The aerosols were collected by fiberglass filters and using a sampling flow of 1. 5 m3 / min. Results: The results show that the largest source of atmospheric particles in Tehran is due to the combustion of fuel released by mobile resources such as transportation and also, stationary resources such as industrial and factories as the second source of atmospheric particles production. Also, mineral particles with irregular shape were observed in coarse particle which may be mainly derived from natural sources such as soil dust, resuspension of dust from road, crust and some other anthropogenic activities such as construction and vehicles. Conclusions: These particles morphology and chemical composition, illustrate an abundance of natural elements within the zone. However, some of the elements presented are directly related to human activities, and are interested by the public health and environmental perspectives.

More than 60% of accidents in Iran occur within 30 kilometers of cities entrance roads. Therefore the number of accidents per kilometer in these regions, in contrast to the other parts of roads is very considerable. The city of Tehran, as the capital of Iran, is the cross point of major arterials of passenger and freight transportation. Thus the evaluation of road safety, entering and exiting the city in the 30 kilometers buffer, can be considered to improve countrywide road safety. To investigate the effects of roadway characteristics and traffic conditions, including number of traffic violations, on accident frequency within 30 km of Tehran entrance roads, Poisson and Negative Binomial regression model were calibrated in this study. Although the goodness of fit index in Poisson model is much greater than the Negative Binomial model, the accident data dispersion in this study implies that Negative Binomial model is more appropriate to show the influences of effective variables. According to results of modeling, the variables including average daily volume, share of heavy vehicles, average daily speed violations and number of direct accesses and interchanges have effective influence on accident frequencies. Controlling access to roads and using effective actions to reduce the speed and number of speed violation can be accounted as effective policies to reduce accident frequencies within 30 km of Tehran entrance roads.

Fast and accurate network traffic identification is becoming essential for network management, high quality of service control and early detection of network traffic abnormalities. Techniques based on statistical features of packet flows have recently become popular for network classification due to the limitations of traditional port and payload based methods. In this paper, we propose a method to identify network traffics. In this method, for cleaning and preparing data, we perform effective preprocessing approach. Then effective features are extracted using the behavioral analysis of application. Using the effective preprocessing and feature extraction techniques, this method can effectively and accurately identify network traffics. For this purpose, two network traffic databases namely UNIBS and the collected database on router are analyzed. In order to evaluate the results, the accuracy of network traffic identification using proposed method is analyzed using machine learning techniques. Experimental results show that the proposed method obtains an accuracy of 97% in network traffic identification.

Analyzing motion patterns in traffic videos can be employed directly to generate high-level descriptions of their content. For traffic videos captured from intersections, usually, we can easily provide additional information about traffic phases. Such information can be obtained directly from the traffic lights or through traffic lights controllers. In this paper, we focus on incorporating additional information to analyze the traffic videos more efficiently. Using side information on traffic phases, the semantic of motion patterns from traffic intersection scenes can be learned more effectively. The learning is performed based on optical flow features extracted from training video clips, and applying them to supervised topic models such as MedLDA and MedSTC. Based on such models, any video clip can be represented based on the learned patterns. Such representations can be further exploited in scene analysis, rule mining, abnormal event detection, etc. Our experiments show that employing side information in intersection video analysis leads to improvement in discovering scene pattern. Moreover, supervised topic models achieve about 4% improvement in abnormal event detection, compared to the unsupervised ones, in terms of area under ROC.