Sharif Digital Repository

DeepFake is a type of forgery that uses deep learning algorithms to make changes to audio and video content that the audience is unable to detect. Nowadays, due to the threats posed by the use of DeepFake to move people's faces in video, researchers' attention has been drawn to designing methods to detect this type of forgery. Detection methods are usually classified into two types. The first case is the extraction of features to detect forgery distortions, for example, the extraction of facial orientations to detect inconsistencies. The second case is the use of deep learning networks for feature extraction and classification, of which the EfficientNet network is an example. Despite the... 

Emotions are one of the most important issues which affects daily life and activities. On the other hand, robots play an increasing role in human life and play a fundamental role in meeting our needs. One of these basic roles is empathy and verbal interaction between the robot and human. In this research, participant's emotions were stimulated in two ways: by using OASIS and GAPED image data sets and by instructing the participants to remind about their good or bad memories. During emotional stimulation, EEG signals have been recorded for the training and testing process. The preprocessing of training data includes filtering, removing bad parts of data, removing bad channels and... 

This thesis investigates the scheduling of television advertisements, which is critical for optimizing the efficiency and profitability of marketing campaigns for TV networks. Since its inception in the 1930s, television has been a cornerstone of advertising, enabling businesses to reach broad and diverse audiences. The evolution of TV advertising strategies, particularly with the rise of digital media competition, underscores the importance of effective advertisement scheduling to maintain relevance and effectiveness. To address this challenge, a comprehensive mixed-integer linear programming (MILP) model was developed. This model incorporates various hypotheses to manage real-world... 

A probabilistic approach to modeling the seismic damage incurred by structural and nonstructural components is developed in this study. For this purpose, Bayesian linear regression is employed to characterize the epistemic uncertainty in the model. In this approach, regressors are constructed based on the mechanics of damage. In turn, the probability distributions of the model parameters and the model error are quantified by Bayesian regression using real-life data from laboratory experiments and/or post-earthquake inspections of damaged components. The study presents the stepwise procedure of collecting the observations, quantifying the repair costs, developing model forms, performing model... 

This thesis works on certain type of underactuated mechanical systems with nonholonomic constrain, namely wheeled robots.Since this robots has different types, this work focuses specifically on three-wheeled robot. At first we know kinematic characteristics of robot system and equations of motion.To control this system, at first we will change equations to chain form, then control it by using known methods. It will be shown that between the known methods for convergence to a fixed point,irregular transformation is the best way for control systems in chain form.Also, with dynamic feedback linearization we can control robot on a trajectory that has a continuous third order derivatives. Then... 

We prove that if an n-dimensional complete minimal submanifold M in hyperbolic space has sufficiently small total scalar curvature then M has only one end. We also prove that for such M there exist no nontrivial L^2 harmonic 1-forms on M  

Taghizadeh Manzari, M. (MehrdadThe main goals of this project are studding the effect of carbon nanotube on the properties of PA6/PA66 matrix and miscibility of PA6/PA66 blend. So PA6/PA66/CNT nanocomposites were compounded by melt mixing method and samples were prepared by injection molding. Due to study miscibility of PA6 and PA66, we used DMTA test and found that PA6 and PA66 are miscible because of existing one pick on DMTA graph. Also DSC test results confirmed that PA6 and PA66 are miscible. Moreover, according to DMTA results, modulus will be rose by increasing CNT content. Mechanical test showed same results in this case. Adding 1%CNT to matrix caused 10% improvement in modulus and... 

Optical computing has emerged as a promising candidate for real-ime and parallel continuous data processing. Optical devices that perform spatial and spatiotemporal transformations of optical signals are of great interest for a wide range of applications including ultrafast all-optical information processing and analog optical computations. Although plasmonic metasurfaces(for analog computing applications) was investigated in the past, they are not realizable. Consequently the silicon photonics is a good alternative for realizing these applications.Firstly, the optical computing and its important features explained and a common architecture was introduced. in addition, another two structures... 

The main goal of this thesis is the design and development of a rehabilitation device for helping patients “to improve blood circulation” & retain and recover their abilities if possible that were lost during the stroke or a accident like that (my research in karaj/iran shows that it is possible by just movement of the specific body part). The focus is on hand in this thesis because Injuries to the hand are common and can be very debilitating since our hands are our primary means for interacting with our world. For rehabilitation in case of human hand there are several practices that should be done for patients either in home or in a rehabilitation center, time and money that is supposed to...