Sharif Digital Repository

The objective of OCR systems is to retrieve machine-encoded text from a raster image. Despite the abundance of powerful OCR algorithms for English, there are not many for Farsi. Our proposed algorithm is comprised of pre-processing, line detection, sub-word detection and segmentation, feature extraction and classification. Furthermore, hardware implementation and acceleration of this system on a GPGPU is presented. This algorithm was tested on 5 fonts including Titr, Lotus,Yekan, Koodak and Nazanin and an average accuracy above 90% was achieved  

Many mobile applications running on smartphones, wearable devices, tiny autonomous robots and IoT devices would potentially benefit from the accuracy and scalability of deep CNN-based machine learning algorithms. However,performance and energy consumption limitations make the execution of such computationally intensive algorithms on embedded mobile devices prohibitive.We present a GPU-accelerated engine, dubbed mCNN, for execution of trained deep CNNs on mobile platforms. The proposed solution takes the trained model as input and automatically optimizes its parallel implementation on the target mobile platform for efficient use of hardware resources such as mobile GPU threads and SIMD units.... 

Spiking Neural Networks(SNN) are networks which are consisted of layers of neurons, like other typical artificial neural networks. The main difference between SNN and other neural networks is the type of data transportation among neurons which is done by spikes. Spiking neural networks and their models are considered as the nearest networks and neurons to animals’ nervous systems. In aspects of hardware implementation, the type of data transportation in SNN causes them to be ultra-low power. So, implementation of these networks on chips like FPGA and also usage of SNN in applications with high processing load have startling germination, recently. In this work, we have tried to propose some... 

Many chip manufacturers have recently introduced high-performance deep-learning hardware accelerators. In modern GPUs, programmable tensor cores accelerate the heavy operations involved in deep neural networks. This paper presents a novel solution to re-purpose tensor cores in modern GPUs for high-throughput implementation of turbo decoders. Turbo codes closely approach Shannon’s limit on channel capacity, and are widely used in many state-of-the-art wireless systems including satellite communications and mobile communications. Experimental evaluations show that the proposed solution achieves about 1.2 Gbps throughput, which is higher compared to previous GPU-accelerated solutions  

There have been many works on generative networks and image generation in the past few years, but the problem with this work is that there is no control over the generated images. The goal of disentangled image synthesis is to generate new images with specific detail and have control over the generated images. Image-based virtual try-on aims to synthesize the customer image with an in-shop clothes image to acquire seamless and natural try-on results, which have attracted increasing attention. The main procedures of image-based virtual try-on usually consist of clothes image generation and try-on image synthesis. In contrast, prior arts cannot guarantee satisfying clothes results when facing... 

In this project, a method is emoloyed to implement a Viterbi decoder on GPGPU. This method is based on combining all steps of the algorithm. This combination has some challenges that are related to differences between different steps of the algorithm. So in this project, some solutions are found to handle these challenges and a high-throughput Viterbi decoder is acheived  

Cardiovascular diseases are the leading cause of death globally so early diagnosis of them is important. Many researchers focused on this field. First signs of cardiac diseases appear in the electrocardiogram signal. This signal represents the electrical activity of the heart so it’s primarily used for the detection and classification of cardiac arrhythmias. Permanent monitoring of this signal is not possible for specialists so we should do this by means of Artificial Intelligence. In this thesis, we use recurrent neural networks to classify electrocardiogram’s arrhythmias. This deep learning method, use two sources of data to learn from. The first part of data is global for everyone and the... 

The extraordinary ability of the human brain to solve various problems has led scientists to simulate models of the human brain. One of these simulated models is artificial neural networks. Today, the power of artificial neural networks is not overlooked. The ability of artificial neural networks to solve various types of issues led us to use the thesis to solve some of the graph-based problems. Quite accurately, this graph-based problem is a matter of identifying the source of rumor in a network. In many graph networks, whether natural networks such as the network of neurons in the human brain or synthetic ones such as the types of social networks, it is possible that a rumor spreads across... 

Real-time simulators are playing a critical role in the design of power electronic systems. Using real-time simulator enables designers to reduce the time and cost of design by employing hardware-in-the-loop technique. In addition, this simulator can potentially reduce the risks which are associated with performing tests on actual systems. In this thesis, a real-time simulator is designed and implemented for power electronic applications. Concerning the fact that power switches are one of the most important elements in this simulator, an ideal switch model with conduction resistor and RC snubber circuit is used. This model is more accurate as compared to other models in the... 

Spiking neural networks (SNNs) are third generation of neural networks. Similar to traditional neural networks, SNNs are comprised neurons. However, not only structure but also information processing is inspired by animal neural systems. SNNs can be called the most similar networks to animal neural systems. In such networks, the information is processed based on propagation of spike signals through the network. The type of data flow in these networks leads to being low-power when they are implemented on hardware. Therefore,there has been a upward trend in hardware implementation of them, like their FPGA implementations, for applications such as Big Data and Machine Learning. In this thesis,... 

The application of deep learning algorithms in the processing of telecommunication signals is increasing. One of the issues in this area is the automatic recognition of radio signal modulation. Recognizing the type of signal modulation as the first stage of baseband signal processing on the receiver side is a key issue in military and civilian applications. The problem with classical algorithms for solving this problem is their strong dependence on channel characteristics and factors resulting from the transmitter and receiver being non-ideal. However, modern algorithms based on deep neural networks have been able to make the model somewhat resistant to these factors. However, the detection... 

Hand pose estimation is something that has applications in many fields, including augmented and virtual reality systems, as well as mixed reality. Hand gesture recognition and classification applications including sign language recognition and non-handheld senarios (such as storefront contactless Survey systems) that have found special cases in the Qovid-19 pandemic period Shows the highest hand pose estimation importance. In our work, we target 2D and 3D estimation at the same time and also use RGB camera as a sensor to record input data. It becomes more economical to achieve Compared with using RGBD or depth sensors . There is only one RGB image in our input. Also there is no contract to... 

Nowadays, Multi-Object Tracking (MOT) plays a crucial role in various computer vision applications such as autonomous vehicles, surveillance, and robotics. Traditional MOT methods often struggle with challenges such as high errors when dealing with complex scenarios involving occlusion, scale variations, and object interactions. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs) and Transformer models, have demonstrated significant capabilities in addressing these challenges. This thesis presents a study on the use of deep learning techniques, specifically CNNs and Transformer models, for solving the problem of Multi-Object Tracking. It begins by... 

Real-time simulators are important in design and test of power systems. Real-time simulators with hardware-in-loop capabilities are used in design of new elements in order to achieve maximum compatibility between subsystems, or in diagnosis and maintenance.The purpose of this dissertation is to design a real-time simulator for electrical machines. For this purpose, a three-phase induction machine is chosen and its model is extracted and discretized. This procedure results in a system of differential equations in form of AX = B. The extracted equations needs to be solved in real-time at every time-step. The real-time algorithms are imple-mented on FPGA. The implemented real-time simulator... 

Nowadays, thanks to improvement of processing hardware and plenty of data available, artificial intelligence and specifically Deep Learning are being one of the powerful tools for solving different problems. Also graph is one of the powerful tools for modeling different data structures. Graph matching is one of the problems in the field of graph problems.In this thesis we consider the problem of graph matching in Erdos-Renyi graphs. The graph matching problem refers to recovering the node-to-node correspondence between two correlated graphs. Previous works theoretically showed that recovering is feasible in sparse Erdos-R´enyi graphs if and only if the probability of having an edge between a... 

The main goal in many fields in the empirical sciences is to discover causal relationships among a set of variables from observational data. PC algorithm is one of the promising solutions to learn underlying causal structure by performing a number of conditional independence tests. In this paper, we propose a novel GPU-based parallel algorithm, called cuPC, to execute an order-independent version of PC. The proposed solution has two variants, cuPC-E and cuPC-S, which parallelize PC in two different ways for multivariate normal distribution. Experimental results show the scalability of the proposed algorithms with respect to the number of variables, the number of samples, and different graph... 

In the communication systems, the raw signals of information are mainly encoded so as to prevent the detrimental effects of channel noises and distortions. After some processes, this encoded signal is passed through the channel. At the receiver side, the received signal has to be decoded to extract the information signal. In order to decode the received signal, the receiver require prior knowledge about the encoder parameters. The traditional approach is to send the encoder parameters along with the encoded signals. However, this transmission overhead might occupy a considerable amount of bandwidth since the type of coding may alter several times in a fraction of a second based on the... 

Positioning refers to the process of estimating the receiver coordinates aiming for an accurate understanding of the surrounding environment. This branch of science has attracted considerable attention in recent years. Today, the influence sphere of positioning is so expanded that encompasses applications in daily life (e.g., navigation) as well as commercial and even military fields. Global Positioning System (GPS) is the most widely used tool in real-time positioning. Because GPS imposes great measurement errors in challenging conditions (e.g., turbulent water environments), alternative methods such as methods based on Wi-Fi, Bluetooth, and visible light have been proposed. Benefiting... 

Cardiovascular diseases are the first leading cause of death in the world also in IRAN. Early detection of such problems can decrease the costs also can help to cure the patient but it needs continuous monitoring and automated classification of hearbeats. Mobile devices and wearable gadgets are good solutions which can help patients before visiting the doctor.In this research, an algorithm is introduced which with the help of ECG signal detects dangerous myocardial problems. Our approach is using deep learning method which were not considered much before. In the proposed algorithm ECG signal is processed in order to get features and with dimensionality reduction, input of the network gets...