Sharif Digital Repository

Meaningful gesture recognition, whose purpose is to interpret human movements, plays a crucial role in various fields such as human and computer interaction, sign language recognition, robot control, and medical applications. Sign language recognition is regarded as the most significant use of gesture recognition by many researchers. Sign languages are the natural medium of communication for millions of deaf people all over the world, and the existence of a sign language recognition system has significantly aided in facilitating communication between deaf individuals and others. Despite numerous studies conducted in this field in recent years, there are still many challenges to continuous... 

Video segmentation means grouping of pixels of the video sequences into spatio-temporal regions which exhibit coherence in both appearance and motion. Due to complexity and spatio-temporal variations, dynamic texture segmentation is a one of the most challenging task in video processing. The problem of dynamic texture segmentation has received considerable attention due to the explosive growth of its applications in video analysis and surveillance systems. In this thesis, two novel approaches have been proposed. The first proposed method is based on generative Dynamic texture models (DTMs) which represent videos as a linear dynamical system. Since DTMs cannot be used for complex videos which... 

World population growth, depletion of fossil fuels resources depletion and increased air pollution resulting from their consumption, human society has had to replace fossil fuels with renewable sources. In this context, the production of hydrogen using solar energy is feasible candidate to replace fossil fuels, which can be produced by water splitting in a photoelectrochemical (PEC) cell using an appropriate photocatalyst. This clean production method has been attended by many researchers in recent years. The most common photocatalyst for the conversion of sunlight to hydrogen by splitting water is Titanium dioxide (TiO2). But high energy gap and fast electron-hole recombination rate in TiO2... 

Software modeling is one of the most important activities in the Software Engineering domain. Models provide an abstract view of the system, thereby facilitating their understandability and resolving the complexities and problems involved. Indeed, models can be considered as a roadmap for achieving the goal of producing software-intensive systems. The aim of real-world modeling is to identify real-world entities, their interactions, and the constituent activities. Real-World Models are tangible to all stakeholders and provide them with an overall view of the Problem Domain. Therefore, many software development methodologies are based on modeling the real world. In some of these... 

In this research an analytical and semi analytical solution to the differential eqautions derived from coupled analysis of heat transfer, moisture transfer and solid deformation in porous materials is presented. In first section, differential equations derived from Hydro-mechanical analysis of an unsaturated soil layer with limited thickness was considered and it is assumed that pore air pressure is equal to atmospheric pressure. To solve the system of equations laplace transform is used.In second section, the effect of daily changes in temperature on transient heat and moisture transfer in semi infinite layer is studied in three dimension. Energy conservation equation and pore liquid mass... 

The ability to diagnose a disease rapidly, non-invasively, with high confidence level and low cost can be of great importance in many ways such as 'detection and on-time conduction of remedial strategies. Hence It can reduce fatality rate and costs in the hygiene field. Currently some pathological diagnosis methods are not suitable considering their precision and costs and in addition to that, most of them are single-function and are developed for diagnosis of a particular disease and don't perform well in the case of multiple disease factors. Many diseases are caused by metabolic disorders and thus direct study of metabolic factors can provide us with a lot of evidence on the disease. One... 

Extracting the minutia of a fingerprint image and then hiding this information in the original image in order to increase the security has been explained and implemented in this thesis. One of the methods which can embed the predetermined digital data in the image is digital watermarking. In this thesis, the host image is a fingerprint image whose minutia has been embedded in that image. Embedding can be done in two spatial and frequency domains. In this thesis, embedding is implemented by combining two domains. Embedding in the frequency domain is applied to transforming coefficients directly from the image then by spatial method the information is embedded. Actually combining DWT and LSB... 

In this study، a novel frequency multiplier chain for wideband millimeter-wave applications at 28 GHz is presented. Millimeter-wave imaging systems require high frequency and wide bandwidth to accurately detect hidden objects. Various methods have been proposed to increase the frequency of generated signals, but in the meantime، frequency multiplier structures exhibit better performance in terms of bandwidth، power consumption and, harmonic rejection. These circuits are widely used in High-frequency telecommunication systems such as 5th generation mobile services. To design a frequency quadrupler، two multiplier stages based on the improved quadrature Gilbert structure were used. Generating... 

Ball bearings diagnosis increases the efficiency and reliability of industrial machinery. Data driven is one of the most effective and highly applicable method in this field. In this way, it is usually assumed that the sensor is mounted on the bearing housing, but in many cases there is no direct access to some of the machine bearings, and in such cases, the bearing can not be monitored and checked. In this project, with simplifications in order to inaccessible bearings diagnostic, a shaft with several supports is considered. The purpose of this project is to diagnose the bearing by the sensors located on the other bearing's housing. In order to achieve this goal, continuous wavelet... 

This study aimed to develop a method for locating damages in the early stages of damage expansion in plate components of the structure using a one-dimensional discrete wavelet function. Continuous monitoring of the health of structures to preserve and repair them to protect the lives and property of individuals and national capitals and to lower the direct and indirect economic costs imposed on society is vital and inevitable for society's decision-makers. Because the damage to the structure caused by earthquakes occurring during the lifetime of the structure, exhaustion of structural components, improper implementation of the structure at the time of construction, and other events that... 

Application of convolution-based transforms such as Fourier transform and wavelet transform are among the widely-used approaches in structural identification problems. These methods, however, are mostly based on a classical damping model for the structures. Hence, such methods cannot be used to accurately identify the modal properties of non-classically damped systems, such as structures equipped with energy dissipated devices, base isolated and multi-segment structures. In this study, complex-valued wavelet transform is used to identify the dynamic properties of structures with arbitrary damping models. This method is known to be able to identify the modal properties of highly damped... 

The failure in rolling element bearings as an important part of rotary machine can lead to machine breakdown.Consequently, bearings can reduce reliability of these components and therefore a maintenance strategy is essential to keep reliability in a desired level. The first and primitive strategy is called breakdown or unplanned maintenance. The next is preventive or planned maintenance. In this scheme, maintenance is done after a specified time period irrespective of machine health status. Out of place maintenance is the disadvantage of preventive maintenance. The most effective generation of maintenance strategy is condition based maintenance (CBM) which suggests maintenance decision based... 

Nowadays high efficiency and power density are desired for designing a power electronic converter. Technology advances in power semiconductor devices, makes it possible to design converters for operation in higher frequencies to reduce size of the converters. As the frequency is increased, switching losses and temperature is also increased, so designing a converter for high frequency operation needs some tradeoffs. LLC resonant converter as a promising solution for battery chargers has ability to adjust its output voltage during Constant Current and Constant Voltage modes. However, its operation mode would change based on switching frequency change. Therefore, for efficiency-oriented... 

In recent years, dialogue generation systems as a branch of natural language text generation have made significant progress. These systems enable human-machine conversations and based on their application, they’re divided into two categories: the goal-oriented, and the open-domain dialogue systems. A goal-oriented dialogue is based on a specific goal, such as the hotel reservation or the product recommendation, while there is no specific topic or goal in the open-domain conversations. This wide coverage of topics in open-domain systems causes some occasional contradictions between utterances. Therefore, preserving dialogue consistency has become a challenge in open-domain dialogue systems.... 

Road accidents are unfortunate events that cause more thanl16000 deaths each year in Iran. Intercity accidents require a comprehensive plan to reduce casualties because the number of roads users are increasing and the accidents account for nearlyl65% of fatalities. In this study, we first tried to identify the status of Iran through a study of traffic accidents in the world, and then the research and activities carried out in Iran were analyzed to find new and effective solutions. Using the daily fatalities data froml2008 tol2014, and using the new methodology presented in this research based on the Discrete Fourier Transformation (DFT), the Box-Jenkins models and the Secant method, the... 

Digital technologies are shaping various industries in the era of the 4th industrial revolution, and meanwhile, we are witnessing the faster development of these technologies in the post-Corona period. In the oil industry, digital fields have created a window of opportunity to provide a competitive advantage for reputable oil companies. Realizing the importance of this window of opportunity, the leading oil companies quickly seek to use their capacities in the development of digital technologies such as automation, artificial intelligence, robotics, twins. Digital and fast processing methods are in the context of a digital field to take a step towards digital transformation. National Iranian... 

One of the most significant factors for the on-time and successful implementation of construction projects is contract management. Proper management of construction contracts and assessment of potential risks in the bidding process and before its signing have a significant impact on preventing or reducing the occurrence of claims and disputes between the contract parties at various stages of the project. In this research, using the latest deep learning (DL) and natural language processing (NLP) state-of-the-art methods, and various deep neural networks (DNN) architectures a model has been developed for extracting Persian contract risk-prone clauses. In addition, this study provides a... 

Pitch or fundamental frequency is one of the components of speech production. Pitch detection is the process of determining the period of the vocal cords closure or in other words, the time duration of one glottal closing, opening and returning phase. There are different algorithms for pitch detection purposes, each of which has some advantages and disadvantages. Pitch detection algorithms can be classified into event pitch detection and non-event pitch detection. Dyadic wavelets transform is an example of event pitch detection. Modified higher order moment, which is based on the Autocorrelation function, is non-event pitch detection. Dyadic wavelets transform is a fast and simple pitch... 

Artificial intelligence (AI) has become an instrumental force in reshaping various sectors, especially in mobile networks. To realize AI as a service (AIaaS) and enable prompt decision-making, edge computing has risen as an innovative response, pushing computation closer to data sources. Although this is promising to reduce the dependency on remote cloud infrastructures and save the mobile devices’ energy, current methodologies display constraints in effectively distributing deep neural network (DNN) inference tasks, leading to possible resource underutilization and performance compromises. To tackle these issues, in this thesis, we study AIaaS at the edge and focus on renowned DNN... 

In this thesis the goal is to use data Mining techniques for finding features of different Partial discharges happen in power transformer insulation defect models in order to monitor the insulation condition of in-service power transformers continuously and on-line and to identify any type of insulation defect in power transformers at the early stage of formation.For this purpose, power transformer Insulation defects are modeled physically and partial discharge current pulses are recorded as Partial discharge data.Tow format of data are investigated, in first one, recorded data in one power frequency cycle are explored and in second one, individual PD pulses are considered for study.In...