Kourosh Kiani
Semnan University, Electrical and Computer Engineering, Faculty Member
Research Interests:
Flat X-ray detectors require a systematic calibration and correction of image artifacts. Based on an analysis of the physics of the image generation chain, this work presents a unified framework for the correction of these artifacts.... more
Flat X-ray detectors require a systematic calibration and correction of image artifacts. Based on an analysis of the physics of the image generation chain, this work presents a unified framework for the correction of these artifacts. Algorithms for the correction steps are presented, including a new method for the calibration and correction of the intertwined offset, gain, and non-linearity as
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
A new approach for embedding the authentication data using fractal and chaos technique is developed in this paper. This approach is to meet the need for authentication, including fragility and inalterability. Using chaos technique and the... more
A new approach for embedding the authentication data using fractal and chaos technique is developed in this paper. This approach is to meet the need for authentication, including fragility and inalterability. Using chaos technique and the keyword extracted from the image itself, it is so unlikely to predict the blocks used for embedding the data. The first section is dedicated
Research Interests:
Research Interests:
Research Interests:
Research Interests: Computer Science, Artificial Intelligence, Sign Language, Sign Languages, American Sign Language, and 15 moreGesture Recognition, Multimodality, Medicine, Deafness and Hearing Loss, Entropy, Learning, Mathematical Sciences, Deep Learning, Physical sciences, Robustness (evolution), Deaf, Boltzmann Machine, Restricted Boltzmann Machine, Aprenentatge, and Convolutional Neural Network
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
One of the demerits of FLC (fuzzy logic controller) is disability in self-tuning which contribute to contingent on knowledge of experts or expert systems. In most cases, tries and errors methodology is used to tune up FLC that could be so... more
One of the demerits of FLC (fuzzy logic controller) is disability in self-tuning which contribute to contingent on knowledge of experts or expert systems. In most cases, tries and errors methodology is used to tune up FLC that could be so time-consuming and may be could not lead to best response. Whereas, metaheuristic algorithms such as Cuckoo Optimization Algorithm (COA) and Particle Swarm Optimization (PSO) could identify the almost optimum parameters of FLC. COA fuzzy controller is one of the most effective methods in term of conditions which designing FLC on account of insufficient expert knowledge is so problematic. There are several controllers approach to this demand but in this paper with the help of COA, a powerful method for tuning fuzzy logic controller is considered and applied for controlling a steam condenser plant. Finally a comparative study between COA-Fuzzy, PSO-Fuzzy and PID controllers is demonstrated to verify the performance of proposed method.
Research Interests:
Research Interests:
Zero-Shot Learning (ZSL) has rapidly advanced in recent years. Towards overcoming the annotation bottleneck in the Sign Language Recognition (SLR), we explore the idea of Zero-Shot Sign Language Recognition (ZS-SLR) with no annotated... more
Zero-Shot Learning (ZSL) has rapidly advanced in recent years. Towards overcoming the annotation bottleneck in the Sign Language Recognition (SLR), we explore the idea of Zero-Shot Sign Language Recognition (ZS-SLR) with no annotated visual examples, by leveraging their textual descriptions. In this way, we propose a multi-modal Zero-Shot Sign Language Recognition (ZS-SLR) model harnessing from the complementary capabilities of deep features fused with the skeleton-based ones. A Transformer-based model along with a C3D model is used for hand detection and deep features extraction, respectively. To make a trade-off between the dimensionality of the skeletonbased and deep features, we use an Auto-Encoder (AE) on top of the Long Short Term Memory (LSTM) network. Finally, a semantic space is used to map the visual features to the lingual embedding of the class labels, achieved via the Bidirectional Encoder Representations from Transformers (BERT) model. Results on four large-scale datas...
Research Interests:
Research Interests:
Sign Language is the dominant form of communication language used in the deaf and hearing-impaired community. To make an easy and mutual communication between the hearing-impaired and the hearing communities, building a robust system... more
Sign Language is the dominant form of communication language used in the deaf and hearing-impaired community. To make an easy and mutual communication between the hearing-impaired and the hearing communities, building a robust system capable of translating the spoken language into sign language and vice versa is fundamental. To this end, sign language recognition and production are two necessary parts for making such a two-way system. Sign language recognition and production need to cope with some critical challenges. In this survey, we review recent advances in Sign Language Production (SLP) and related areas using deep learning. To have more realistic perspectives to sign language, we present an introduction to the Deaf culture, Deaf centers, psychological perspective of sign language, the main differences between spoken language and sign language. Furthermore, we present the fundamental components of a bi-directional sign language translation system, discussing the main challenge...
Research Interests:
Sign Language Recognition (SLR) is a challenging research area in computer vision. To tackle the annotation bottleneck in SLR, we formulate the problem of Zero-Shot Sign Language Recognition (ZS-SLR) and propose a two-stream model from... more
Sign Language Recognition (SLR) is a challenging research area in computer vision. To tackle the annotation bottleneck in SLR, we formulate the problem of Zero-Shot Sign Language Recognition (ZS-SLR) and propose a two-stream model from two input modalities: RGB and Depth videos. To benefit from the vision Transformer capabilities, we use two vision Transformer models, for human detection and visual features representation. We configure a transformer encoder-decoder architecture, as a fast and accurate human detection model, to overcome the challenges of the current human detection models. Considering the human keypoints, the detected human body is segmented into nine parts. A spatio-temporal representation from human body is obtained using a vision Transformer and a LSTM network. A semantic space maps the visual features to the lingual embedding of the class labels via a Bidirectional Encoder Representations from Transformers (BERT) model. We evaluated the proposed model on four dat...
Congestion and overloading for lines are the main problems in the exploitation of power grids. The consequences of these problems in deregulated systems can be mentioned as sudden jumps in prices in some parts of the power system, lead to... more
Congestion and overloading for lines are the main problems in the exploitation of power grids. The consequences of these problems in deregulated systems can be mentioned as sudden jumps in prices in some parts of the power system, lead to an increase in market power and reduction of competition in it. FACTS devices are efficient, powerful and economical tools in controlling power flows through transmission lines that play a fundamental role in congestion management. However, after removing congestion, power systems due to targeting security restrictions may be managed with a lower voltage or transient stability rather than before removing. Thus, power system stability should be considered within the construction of congestion management. In this paper, a multi-objective structure is presented for congestion management that simultaneously optimizes goals such as total operating cost, voltage and transient security. In order to achieve the desired goals, locating and sizing of series ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Speeding up the system is one of the basic challenges in the real-world applications of Face Recognition (FR), whereas reducing the computational complexity can significantly increase the speed of the system. In recent years, many face... more
Speeding up the system is one of the basic challenges in the real-world applications of Face Recognition (FR), whereas reducing the computational complexity can significantly increase the speed of the system. In recent years, many face recognition methods have been proposed but few of them give attention to this issue. Accordingly, in this article, we take the axis-symmetrical property of faces as a novel idea to speed up the face recognition algorithm as well as to reduce the computational complexity. Taking the axis-symmetrical property of faces leads us to use half of the face image. Proposing a face recognition system using Hidden Markov Model (HMM) as a classifier, we use the Singular Value Decomposition (SVD) to build the observation vectors. Evaluated results of the proposed system on Yale and Faces94 datasets show that the proposed system can achieve a satisfactory recognition rate with a higher speed.
Research Interests:
Abstract: Trial and error method can be used to find a suitable design of a fuzzy controller. Generally, the design of fuzzy controller involves determination of the fuzzy rules, Membership Functions (MFs) and scaling factors. An... more
Abstract: Trial and error method can be used to find a suitable design of a fuzzy controller. Generally, the design of fuzzy controller involves determination of the fuzzy rules, Membership Functions (MFs) and scaling factors. An optimization algorithm facilitates the design process and finds an optimal design to achieve a desired performance. This paper presents an Improved Bacterial Foraging Optimization Algorithm (IBFOA) to design a fuzzy controller for tracking control of a PUMA 560 robot arm driven by permanent magnet DC motors. We use efficiently the IBFOA to form the rule base and MFs. To show the improvement of proposed algorithm, the IBFOA is compared with Bacterial Foraging Optimization Algorithm (BFOA) and Particle Swarm Optimization (PSO) algorithm. Performance of the controller in the joint space and in the Cartesian space is evaluated. Simulation results show superiority of the IBFOA to the BFOA and PSO algorithm.
Research Interests: Engineering and Key words
Summary Integration of wind resources into the power systems has increased the technical and financial concerns in transmission expansion planning. In this paper, a stochastic structure is demonstrated for transmission expansion planning... more
Summary Integration of wind resources into the power systems has increased the technical and financial concerns in transmission expansion planning. In this paper, a stochastic structure is demonstrated for transmission expansion planning with allocation of fixed series compensation under wind and load uncertainties. Fixed series compensations have the ability to increase the transfer capacity of transmission lines. However, their significance benefit for transmission expansion planning is their higher effectiveness in power dispatching, which results in lower investment costs, compared with planning without fixed series compensations. The proposed planning model simultaneously optimizes targets such as the investment cost, the congestion cost, and the expected energy not supplied in each stage. The presented planning model is a complicated nonlinear optimizing problem. For introducing group of nondominated optimal solutions, multi-objective gray wolf optimizer algorithm based on pro...
Image restoration and its different variations are important topics in low-level image processing. One of the main challenges in image restoration is dependency of current methods to the corruption characteristics. In this paper, we have... more
Image restoration and its different variations are important topics in low-level image processing. One of the main challenges in image restoration is dependency of current methods to the corruption characteristics. In this paper, we have proposed an image restoration architecture that enables us to address different types of corruption, regardless of type, amount and location. The main intuition behind our approach is restoring original images from abstracted perceptual features. Using an encoder-decoder architecture, image restoration can be defined as an image transformation task. Abstraction of perceptual features is done in the encoder part of the model and determines the sampling point within original images' Probability Density Function (PDF). The PDF of original images is learned in the decoder section by using a Generative Adversarial Network (GAN) that receives the sampling point from the encoder part. Concretely, sampling from the learned PDF restores original image fr...
Research Interests:
Congestion and overloading for lines are the main problems in the exploitation of power grids. The consequences of these problems in deregulated systems can be mentioned as sudden jumps in prices in some parts of the power system, lead to... more
Congestion and overloading for lines are the main problems in the exploitation of power grids. The consequences of these problems in deregulated systems can be mentioned as sudden jumps in prices in some parts of the power system, lead to an increase in market power and reduction of competition in it. FACTS devices are efficient, powerful and economical tools in controlling power flows through transmission lines that play a fundamental role in congestion management. However, after removing congestion, power systems due to targeting security restrictions may be managed with a lower voltage or transient stability rather than before removing. Thus, power system stability should be considered within the construction of congestion management. In this paper, a multi-objective structure is presented for congestion management that simultaneously optimizes goals such as total operating cost, voltage and transient security. In order to achieve the desired goals, locating and sizing of series ...
Research Interests:
A variety of approaches have been proposed for addressing different image restoration challenges. Recently, deep generative models were one of the mostly used ones. In this paper, a new Restricted Boltzmann Machines (RBM) training... more
A variety of approaches have been proposed for addressing different image restoration challenges. Recently, deep generative models were one of the mostly used ones. In this paper, a new Restricted Boltzmann Machines (RBM) training algorithm for addressing corrupted data has been proposed. RBMs can be trained both supervised and unsupervised, however they are very sensitive to noise and occlusion. The proposed algorithm enables the RBM to be robust against corruptions. Using the new algorithm, we have given the RBM a posterior knowledge about desired or clean data. Despite other methods, the proposed algorithm works fine without changing the architecture of the model or adding any regularization term. Concretely, the RBM can be used as a robust feature extractor, even for unclean data. By creating different corrupted versions for each image instance, and using the original version in the reconstruction phase, the RBM can learn the desired probability distribution of data. Experimenta...
Research Interests:
In this paper, we propose a novel approach for image classification based on Graph-based image segmentation method and apply it on SAR images with satisfactory clustering performance and low computational cost. In this method first, the... more
In this paper, we propose a novel approach for image classification based on Graph-based image segmentation method and apply it on SAR images with satisfactory clustering performance and low computational cost. In this method first, the image pre-processes by mean shift algorithm to cluster into disjoint region, then the segmented regions are represented as a graph structure with all connected neighbourhood, and after that normalized cut method is applied to classify image into defined classes. Streszczenie. W artykule przedstawiono metode klasyfikacji obrazow, z wykorzystaniem segmentacji metodą grafową. Proponowana rozwiązanie wykorzystano w analizie obrazow SAR (ang. Specific Absorption Rate), uzyskując dobrą skutecznośc i niski koszt obliczeniowy. (Segmentacja metodą grafową w klasyfikacji obrazow w zastosowaniu do obrazow SAR).
Research Interests:
Image colorization is an interesting yet challenging task due to the descriptive nature of getting a natural-looking color image from any grayscale image. To tackle this challenge and also have a fully automatic procedure, we propose a... more
Image colorization is an interesting yet challenging task due to the descriptive nature of getting a natural-looking color image from any grayscale image. To tackle this challenge and also have a fully automatic procedure, we propose a Convolutional Neural Network (CNN)-based model to benefit from the impressive ability of CNN in the image processing tasks. To this end, we propose a deep-based model for automatic grayscale image colorization. Harnessing from convolutional-based pre-trained models, we fuse three pre-trained models, VGG16, ResNet50, and Inception-v2, to improve the model performance. The average of three model outputs is used to obtain more rich features in the model. The fused features are fed to an encoder-decoder network to obtain a color image from a grayscale input image. We perform a step-by-step analysis of different pre-trained models and fusion methodologies to include a more accurate combination of these models in the proposed model. Results on LFW and Image...
Research Interests:
باختنا اهرازبا .دندش ی هزادنا یریگ هس لماش شسرپ مان ة شیارگ ،درادناتسا ب ،ترجاهم ه شسرپ مان ة هاگشناد وج و شسرپ مان ة هنیمز یبای ( زتراوش یشزرا SVS ) دوب اپ هک اهنآ ییای اب شور بیرض ب خابنورک یافلآ .دش دروآر هتفای یاه جاهم هب شیارگ نازیم... more
باختنا اهرازبا .دندش ی هزادنا یریگ هس لماش شسرپ مان ة شیارگ ،درادناتسا ب ،ترجاهم ه شسرپ مان ة هاگشناد وج و شسرپ مان ة هنیمز یبای ( زتراوش یشزرا SVS ) دوب اپ هک اهنآ ییای اب شور بیرض ب خابنورک یافلآ .دش دروآر هتفای یاه جاهم هب شیارگ نازیم شهوژپ نیب رد تر یوجشناد نا هب طسوتم ار لااب داد ناشن . نینچمه هاگشناد وج داعبا یمامت ، و اظن یشزرا م ، ض نم ییاناوت ،هطبار نتشاد شیپ ینیب .دنتشاد ار نایوجشناد ترجاهم هب شیارگ نب نیاربا اب ب یسرر ریثأت یایاوز نوگانوگ هاگشناد وج و نایوجشناد یشزرا ماظن ، یم ناوت هب یگنوگچ راتفر ، تاساسحا ، ید هاگد اه و شرگن نانآ صوصخرد شیارگ هب ترجاهم یپ درب و شنکاو حا یلامت نانآ ار یبایزرا ، شیپ ینیب و یتح تیاده درک . یلک هژاود اه : یشزرا ماظن ،هاگشناد وج ،ترجاهم هب شیارگ ،اهزغم رارف .
Research Interests:
In this research a multi-objective Imperialist Competitive Algorithm (MOICA) is applied for Environmental and Economic Power Dispatch (EED) problem. Due to the environmental concerns that arise from the emissions produced via... more
In this research a multi-objective Imperialist Competitive Algorithm (MOICA) is applied for Environmental and Economic Power Dispatch (EED) problem. Due to the environmental concerns that arise from the emissions produced via fossil-fueled electric power plants, the classical economic dispatch, which operates electric power systems so as to minimize only the total fuel cost, can no longer be considered alone. Hence, by environmental dispatch, emissions can be reduced by dispatch of power generation to minimize emissions. Also, power generated, system loads, fuel cost and emission coefficients are subjected to inaccuracies and uncertainties in real-world situations. The proposed technique has been carried out on the IEEE 30-bus and 118-bus test system. The results demonstrate the capability of the proposed MOICA approach to solve of multi-objective EED problem. The comparison reported results with MODE and other techniques reveals the superiority of the proposed MOICA approach and co...
Research Interests:
textabstractA large number of people with a movement problem forms a relevant social and medical problem in all countries. The rapidly growing number of elderly people. who inevitably experience increasing limitations in their functioning... more
textabstractA large number of people with a movement problem forms a relevant social and medical problem in all countries. The rapidly growing number of elderly people. who inevitably experience increasing limitations in their functioning as they grow older. is a cause of major international concern. Only in the European Community. 10% of the population is suffering from more or less severe motor problems. Awareness of disability costs and demographic developments have directed the poHcy of goverrunents to quality of life problems. More than in the past, research devoted to diseases of the neuro·musculoskeletal system is supported. This regards diagnosis. surgical and non-surgical treatment, rehabilitation and prevention. In all of these areas biomechanics is essential for the assessment of the mechanical functioning of healthy subjects and patients. Movement analysis is one of the most important parts of biomechanlcal research. Since the end of the 19th century there have been atte...
Research Interests:
A b s t r a c t This paper presents a new artificial intelligence based approach to solve problems in Economic Dispatch (ED) such as total fuel cost that may be introduced in smooth and non-smooth form. This new artificial intelligence... more
A b s t r a c t This paper presents a new artificial intelligence based approach to solve problems in Economic Dispatch (ED) such as total fuel cost that may be introduced in smooth and non-smooth form. This new artificial intelligence algorithm is called Artificial Fish Swarm Algorithm (AFSA). At the first, this new algorithm introduced and the way which this algorithm can be used to solve economic dispatch problems are explained. In order to show that this new algorithm can easily solve this problem and to prove that the use of this new algorithm is more efficient, AFSA has been run for complex problem that consists of non-smooth cost function such as valve-point discontinuities in compared to results of other algorithms in published papers. It is observed that for any kind of problems in ED, this algorithm can be used and best results can be obtained.
Dealing with disguises, illumination and expression variations are important and challenging problems in the face recognition area. Considering the axis-symmetrical structure of the face, we propose a face recognition algorithm using the... more
Dealing with disguises, illumination and expression variations are important and challenging problems in the face recognition area. Considering the axis-symmetrical structure of the face, we propose a face recognition algorithm using the ergodic Hidden Markov Model (HMM) as a classifier and the image of half of the face; while the Discrete Wavelet Transform (DWT) is applied to generate observation vectors. We evaluate the proposed method on AR, Yale and Faces94 face datasets. The results represent the superiority of our method, compared with some other state-of-the-art methods, in terms of recognition rates, computational complexity, and memory consumption.
This study presents a method to reconstruct a high-resolution image using a deep convolution neural network. We propose a deep model, entitled Deep Block Super Resolution (DBSR), by fusing the output features of a deep convolutional... more
This study presents a method to reconstruct a high-resolution image using a deep convolution neural network. We propose a deep model, entitled Deep Block Super Resolution (DBSR), by fusing the output features of a deep convolutional network and a shallow convolutional network. In this way, our model benefits from high frequency and low frequency features extracted from deep and shallow networks simultaneously. We use the residual layers in our model to make repetitive layers, increase the depth of the model, and make an end-to-end model. Furthermore, we employed a deep network in up-sampling step instead of bicubic interpolation method used in most of the previous works. Since the image resolution plays an important role to obtain rich information from the medical images and helps for accurate and faster diagnosis of the ailment, we use the medical images for resolution enhancement. Our model is capable of reconstructing a high-resolution image from low-resolution one in both medica...
Research Interests:
Sign Language is the dominant yet non-primary form of communication language used in the deaf and hearing-impaired community. To make an easy and mutual communication between the hearing-impaired and the hearing communities, building a... more
Sign Language is the dominant yet non-primary form of communication language used in the deaf and hearing-impaired community. To make an easy and mutual communication between the hearing-impaired and the hearing communities, building a robust system capable of translating the spoken language into sign language and vice versa is fundamental. To this end, sign language recognition and production are two necessary parts for making such a two-way system. Sign language recognition and production need to cope with some critical challenges. In this survey, we review recent advances in Sign Language Production (SLP) and related areas using deep learning. This survey aims to briefly summarize recent achievements in SLP, discussing their advantages, limitations, and future directions of research.