Rahmati Mohammad

Rahmati Mohammad
Amirkabir University of Technology | TUS · Department of Computer Engineering and Information Technology

About

119
Publications
19,123
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,558
Citations

Publications

Publications (119)
Preprint
Full-text available
The reliability of a learning model is key to the successful deployment of machine learning in various industries. Creating a robust model, particularly one unaffected by adversarial attacks, requires a comprehensive understanding of the adversarial examples phenomenon. However, it is difficult to describe the phenomenon due to the complicated natu...
Preprint
Over the past few decades, numerous attempts have been made to address the problem of recovering a high-resolution (HR) facial image from its corresponding low-resolution (LR) counterpart, a task commonly referred to as face hallucination. Despite the impressive performance achieved by position-patch and deep learning-based methods, most of these t...
Preprint
Full-text available
Over the past few decades, numerous attempts have been made to address the problem of recovering a high-resolution (HR) facial image from its corresponding low-resolution (LR) counterpart, a task commonly referred to as face hallucination. Despite the impressive performance achieved by position-patch and deep learning-based methods, most of these t...
Preprint
Full-text available
Video Generation is a relatively new and yet popular subject in machine learning due to its vast variety of potential applications and its numerous challenges. Current methods in Video Generation provide the user with little or no control over the exact specification of how the objects in the generate video are to be moved and located at each frame...
Preprint
Full-text available
In this paper, we study the adversarial examples existence and adversarial training from the standpoint of convergence and provide evidence that pointwise convergence in ANNs can explain these observations. The main contribution of our proposal is that it relates the objective of the evasion attacks and adversarial training with concepts already de...
Preprint
Full-text available
In this paper, we propose a novel pooling layer for graph neural networks based on maximizing the mutual information between the pooled graph and the input graph. Since the maximum mutual information is difficult to compute, we employ the Shannon capacity of a graph as an inductive bias to our pooling method. More precisely, we show that the input...
Article
In this paper, a new discriminative dictionary learning algorithm is introduced. An entropy based criterion is embedded into the objective function to enforce a proper structure for the dictionary items when decomposing signals of different classes. The proposed criterion influences the dictionary items to participate in the decomposition of a smal...
Conference Paper
In this paper, we study the adversarial examples existence and adversarial training from the standpoint of convergence and provide evidence that pointwise convergence in ANNs can explain these observations. The main contribution of our proposal is that it relates the objective of the evasion attacks and adversarial training with concepts already de...
Article
Full-text available
Recommender systems provide personalized recommendations to the users from a large number of possible options in online stores. Matrix factorization is a well-known and accurate collaborative filtering approach for recommender system, which suffers from cold-start problem for new users and items. When new users join the system, it will take some ti...
Preprint
Full-text available
Abnormal event detection (AED) in urban surveillance videos has multiple challenges. Unlike other computer vision problems, the AED is not solely dependent on the content of frames. It also depends on the appearance of the objects and their movements in the scene. Various methods have been proposed to address the AED problem. Among those, deep lear...
Article
Full-text available
Glioblastoma (GBM) is the commonest primary malignant brain tumor in adults, and despite advances in multi-modality therapy, the outlook for patients has changed little in the last 10 years. Local recurrence is the predominant pattern of treatment failure, hence improved local therapies (surgery and radiotherapy) are needed to improve patient outco...
Article
Full-text available
The sharing economy, which is empowered by digital technologies, drives platform business models and creates new forms of growth and value, which disrupts many businesses. The essential challenge for the formation of business interactions between individuals in the context of platforms is mutual trust. Blockchain, with broad applications in differe...
Article
In this study, we consider the problem of subspace clustering in the presence of spatially contiguous noise, occlusion, and disguise. We argue that self-expressive representation of data, which is a key characteristic of current state-of-the-art approaches, is severely sensitive to occlusions and complex real-world noises. To alleviate this problem...
Preprint
Recommender systems provide personalized recommendations to the users from a large number of possible options in online stores. Matrix factorization is a well-known and accurate collaborative filtering approach for recommender system, which suffers from cold-start problem for new users and items. Whenever a new user participate with the system ther...
Article
Full-text available
In this paper we present a model-based image steganography method in discrete wavelet transform (DWT). This method is based on the human visual system model. The proposed steganography method assumes a model for cover image statistics. In this algorithm, the DWT coefficients are used as the carrier of the hidden message. An unpleasant outcome of th...
Preprint
In this paper, we consider the problem of subspace clustering in presence of contiguous noise, occlusion and disguise. We argue that self-expressive representation of data in current state-of-the-art approaches is severely sensitive to occlusions and complex real-world noises. To alleviate this problem, we propose a hierarchical framework that brin...
Article
In this paper, a new framework is presented to enhance the reconstruction and discrimination capabilities of existing discriminative dictionary learning methods. In the proposed framework, a non-linear mapping model is introduced to learn a feature space in a way that any standard discriminative dictionary learning algorithms could achieve higher c...
Article
In this paper, we consider the problem of subspace clustering for image data under occlusion and gross spatially contiguous noise. The state of the art subspace clustering methods assume that the noise either follows independent Laplacian or Gaussian distributions. However, the realistic noise is much more complicated and exhibits different structu...
Article
The generalization and robustness of an electroencephalogram (EEG)-based system are crucial requirements in actual practices. To reach these goals, we propose a new EEG representation that provides a more realistic view of brain functionality by applying multi-instance (MI) framework to consider the non-stationarity of the EEG signal. In this repre...
Article
Objective: The generalization and robustness of an electroencephalogram (EEG)-based system are crucial requirements in actual practices. Approach: To reach these goals, we propose a new EEG representation that provides a more realistic view of brain functionality by applying multi-instance (MI) framework to consider the non-stationarity of the E...
Article
Full-text available
In this paper, we focus mainly on designing a Multi-Target Object Tracking algorithm that would produce high-quality trajectories while maintaining low computational costs. Using online association, such features enable this algorithm to be used in applications like autonomous driving and autonomous surveillance. We propose CNN-based, instead of ha...
Article
Full-text available
Sparse subspace clustering (SSC) is one of the current state-of-the-art method for partitioning data points into the union of subspaces, with strong theoretical guarantees. However, it is not practical for large data sets as it requires solving a LASSO problem for each data point, where the number of variables in each LASSO problem is the number of...
Article
Full-text available
The goal of this paper is to present a new efficient image segmentation method based on evolutionary computation which is a model inspired from human behavior. Based on this model, a four layer process for image segmentation is proposed using the split/merge approach. In the first layer, an image is split into numerous regions using the watershed a...
Article
Full-text available
We propose a novel couple mappings method for low resolution face recognition using deep convolutional neural networks (DCNNs). The proposed architecture consists of two branches of DCNNs to map the high and low resolution face images into a common space with nonlinear transformations. The branch corresponding to transformation of high resolution i...
Article
In low rank approximation methods it is often assumed that the data matrix is composed of two globally low rank and sparse matrices. Moreover, real data matrices often consist of local patterns in multiple scales. The conventional low rank approximation techniques do not reveal the local patterns from the data matrices. This paper presents an appro...
Article
Full-text available
Learning Bayesian network structures from data is known to be hard, mainly because the number of candidate graphs is super-exponential in the number of variables. Furthermore, using observational data alone, the true causal graph is not discernible from other graphs that model the same set of conditional independencies. In this paper, it is investi...
Conference Paper
Copy-move is a simple and effective operation for creating digital image forgeries, where an area of an image is copied and pasted to a different location in that image. Generally, a forger uses some affine transformations to make the changes visually intact. Most existing copy-move detection methods are not effective when copied regions are under...
Article
In general, a multi-object tracking problem in a network of cameras consists of two steps. First, objects are tracked in each camera and tracklets, which are the traces of each object in each camera, are extracted by any common tracking algorithm. In the second step, the extracted tracklets are associated and persistent trace of the objects are obt...
Article
In this paper, a novel variational method is introduced for multi-object tracking in a network of cameras. In a camera network, objects are tracked by each camera using any of conventional algorithms and their tracks are extracted. Each extracted track is called a tracklet. The extracted tracklets are the inputs to our proposed method. Our objectiv...
Article
Various ensemble methods are proposed to aggregate hierarchical clusterings. These methods combine a set of hierarchical clusterings into a single representative clustering with an improved quality. The quality of this representative hierarchical clustering intensively depends on the aggregation operator (aggregator) used in the combination. Howeve...
Article
In many wide area surveillance applications, tracking objects is usually accomplished by using network of cameras. A common approach to any multi-objects tracking algorithm in a network of cameras comprises of two main steps. First, the movement trajectory of each object, within the field of view of a camera, is extracted and is called object track...
Article
An important property of any robust steganographic method is that it must introduce minimal distortion in the created stego-images. This objective is achieved if one can maximize the similarity between the pixels value of the cover image and the secret data. In the proposed framework, the maximal similarity is obtained by arranging some routes alon...
Article
Some of the basic algorithms for learning the structure of Bayesian networks, such as the well-known K2 algorithm, require a prior ordering over the nodes as part of the input. It is well known that the accuracy of the K2 algorithm is highly sensitive to the initial ordering. In this paper, we introduce the aggregation of ordering information provi...
Article
Full-text available
In this paper, we propose a new framework for view independent action recognition, which uses a combination of a view-dependent representation and a view-independent representation. The view-dependent representation reduces the number of possible action’s labels prior to the view-independent representation. We used the entropy of silhouette’s dista...
Article
Recently, multi-stable Neural Networks (NN) with exponential number of attractors have been presented and analyzed theoretically; however, the learning process of the parameters of these systems while considering stability conditions and specifications of real world problems has not been studied. In this paper, a new class of multi-stable NNs using...
Article
In Hierarchical clustering, a set of patterns are partitioned into a sequence of groups represented as a dendrogram. The dendrogram is a tree representation where each node is associated with merging of two (or more) partitions and hence each partition is nested into the next partition. Hierarchical representation has properties that are useful for...
Article
Manifold learning algorithms do not extract the structure of datasets in an abstract form or they do not have high performance for complex data.In this paper, a method for Learning an Inductive Riemannian Manifold in Abstract form (LIRMA) is presented in which the structure of patterns is determined by solving the embedded dynamical system of the p...
Article
A genetic programming (GP) algorithm is developed to estimate the minimum spouting velocity (U-ms) in the spouted beds with a cone base. In order to have a general model, five dimensionless variables including seven critical geometric and operating parameters of spouted beds, namely, column diameter, spout nozzle diameter, base angle, static bed he...
Article
Full-text available
This paper presents a model-based image steganography method based on Watson’s visual model. Model-based steganography assumes a model for cover image statistics. This approach, however, has some weaknesses, including perceptual detectability. We propose to use Watson’s visual model to improve perceptual undetectability of model-based steganography...
Article
In many supervised learning problems, determining the true labels of training instances is expensive, laborious, and even practically impossible. As an alternative approach, it is much easier to collect multiple subjective (possibly noisy) labels from human labelers, especially with the crowdsourcing services such as Amazon’s Mechanical Turk. The c...
Article
A recent trend in video coding is toward the low-complexity distributed techniques which provide an adaptive way to distribute the computational complexity among the encoder(s) and the decoder. One of the well-known architectures for Distributed Video Coding (DVC) is the Stanford architecture. This structure imposes the presence of a feedback chann...
Article
A complex network provides a framework for modeling many real-world phenomena in the form of a network. In general, a complex network is considered as a graph of real world phenomena such as biological networks, ecological networks, technological networks, information networks and particularly social networks. Recently, major studies are reported f...
Article
Classification of data is an important problem which has attracted many researchers to introduce new approaches. In this paper, we propose Mapping to Optimal Regions (MOR) as a new method for multi-class classification task to reduce computational and memory complexities. It requires only one simple mapping from input space to optimal regions. The...
Article
Full-text available
In Hierarchical Clustering (HC), a set of patterns is partitioned into a sequence of groups and the results are represented by a dendrogram. The dendrogram is a tree representation where each node is associated with merging of two (or more) partitions and hence each partition is nested into the next partition. Hierarchical representation has valuab...
Article
Mapping to Multidimensional Optimal Regions ((MOR)-O-2) is the enhanced version of Mapping to Optimal Regions (MOR) which is a special purposed method for multiclass classification task. Similar to MOR, it reduces computational complexity; however, presents better accuracy. Theoretical and experimental results confirm that by using (MOR)-O-2, the m...
Conference Paper
Considering into account the most important requirement of steganography, i.e. producing the minimal distortion, the proposed method changes the arrangement of the pixels of the cover image and finds a route such that the maximal similarity between the LSB of the pixels along this route and the secret data is obtained. In the introduced image stega...
Article
Full-text available
This work presents adaptive image steganography methods which locate suitable regions for embedding by contourlet transform, while embedded message bits are carried in discrete cosine transform coefficients. The first proposed method utilizes contourlet transform coefficients to select contour regions of the image. In the embedding procedure, some...
Conference Paper
In this paper a sparse coding approach is proposed. Due to the similarity of the frequency and orientation representations of Gabor filters and those of the human visual system, we have used Gabor filters in the step of creating the dictionary. It has been shown that not all Gabor filters in a typical Gabor bank is necessary and efficient in facial...
Conference Paper
During the last decades a large set of video archives is created and rapidly multimedia growth creates new challenge in the image processing world. A reliable system is needed to automate the process of this large amount of data. Video analyses are done in two different levels, low level and high level. There are many problems in video content anal...
Conference Paper
Reducing Computational complexity is a major issue in data mining. Mapping to Multidimensional Optimal Regions (M2OR) is a special purposed method for multiclass classification task. It reduces computational complexity in comparison to the other concepts of classifiers. In this paper, the accuracy of M2OR increases using Learning Inductive Riemanni...
Conference Paper
In this study a method for hand gesture recognition using dynamic Bayesian networks was presented. This study includes two main subdivisions namely: hand posture recognition and dynamic hand gesture recognition (without hand posture recognition). In the first session, after hand segmentation using a method based on histogram of direction and fuzzy...
Article
This paper introduces the modular cellular neural network (CNN), which is a new CNN structure constructed from nine one-layer modules with intercellular interactions between different modules. The new network is suitable for implementing many image processing operations. Inputting an image into the modules results in nine outputs. The topographic c...
Article
Full-text available
Text in scene images can provide useful and vital information for content-based image analysis. Therefore, localization of text in images is an important task. In this paper, we present a hybrid approach to localize Farsi text in natural scene images. Complex background, variations of text font, size and line orientation and non-uniform illuminatio...
Conference Paper
Nowadays providing people's safety in public places is an important issue for governments and security organizations. Anomaly detection in surveillance videos is one of the applications which helps to manage these issues automatically. Particularly, anomaly detection in crowded scenes such as airports, rail stations and etc has attracted a lot of r...
Conference Paper
Universal steganalysis detect the presence of secret messages. In this paper, we proposed a new methodology which will be created different groups of images based on image features as an unsupervised learning and proposed methodology can enable us to design specific models of steganalyzer in order to improve system accuracy. These models will be di...
Conference Paper
In this paper, we proposed a new statistical framework for blind image steganalysis that is shown to be of higher detection performance accuracy than truly current steganalysis systems. Therefore, we have introduced a multi-classification methodology based on image features to group images into the optimal classes in order, to make the models speci...
Conference Paper
Distributed Video Coding (DVC) is a new class of video coding techniques with the aim of coding the decentralized video sources. While the Stanford Wyner-Ziv codec is a well-known architecture in DVC literature, one of its main drawbacks is the presence of a feedback channel from the decoder to the encoder. This feedback channel makes the use of th...
Conference Paper
Full-text available
Compressive Sensing (CS) is a new method for sparse images reconstruction using incomplete measurements. In this study our goal is to reconstruct a High Resolution (HR), MR image from a single Low Resolution (LR) image. Our proposed method applies the CS theory to Super Resolution (SR) single Magnetic Resonance Imaging (MRI). We first use a LR imag...
Conference Paper
In this paper we proposed a new method for the problem of structural human action recognition in single images. In this work, we first extract all Poselets in the images for using as the descriptor of human's activity. Then, we model the latent topics of human poses by using extracted vectors and P-LSA. Finally recognize human's action in a query i...
Article
In this paper, a new wave computing algorithm for edge detection in real images is introduced. This algorithm is suitable for real time applications due to the parallel processing capabilities of CNN. The new algorithm is based on the wave computing concept, using diffusion for noise reduction and weak edge elimination and trigger wave to emphasize...
Article
Mapping to Multidimensional Optimal Regions (M2OR) is a special purposed method for multiclass classification task. It reduces computational complexity in comparison to the other concepts of classifiers. In order to increase the accuracy of M2OR, its code assignment process is enriched using PCA. In addition to the increment in accuracy, correspond...
Article
Full-text available
Three new learning algorithms for Takagi-Sugeno-Kang fuzzy system based on training error and genetic algorithm are proposed. The first two algorithms are consisted of two phases. In the first phase, the initial structure of neuro-fuzzy network is created by estimating the optimum points of training data in input-output space using KNN (for the fir...
Conference Paper
An important application of machine vision and image processing could be driver drowsiness detection system due to its high importance. In recent years there have been many research projects reported in the literature in this field. In this paper, unlike conventional drowsiness detection methods, which are based on the eye states alone, we used fac...
Article
Constant working surveillance cameras in public places, such as airports and banks, produce huge amount of video data. Faces in such videos can be extracted in real time. However, most of these detected faces are either redundant or useless. Redundant information adds computational costs to facial analysis systems and useless data makes the final r...
Conference Paper
Full-text available
To detect and track eye images with complex background, distinctive features of user eye are used. Generally, an eye-tracking and detection system can be divided into four steps: Face detection, eye region detection, pupil detection and eye tracking. To find the position of pupil, first, face region must be separated from the rest of the image usin...
Article
Although current blind image steganalysis systems utilize a wide variety of features and classifiers, a common shortcoming in all of them is that they almost have similar processes for all images and they do not take advantage of the content diversity of different images. In this paper, a new framework is proposed that enables us to employ the cont...
Article
Full-text available
Clustering-combination methods have received considerable attentions in recent years, and many ensemble-based clustering methods have been introduced. However, clustering-combination techniques have been limited to ??flat?? clustering combination, and the combination of hierarchical clusterings has yet to be addressed. In this paper, we address and...
Article
In this paper, we used data mining techniques for the automatic discovering of useful temporal abstraction in reinforcement learning. This idea was motivated by the ability of data mining algorithms in automatic discovering of structures and patterns, when applied to large data sets. The state transitions and action trajectories of the learning age...
Article
A video stream is usually massive in terms of data content with abundant information. In the past, extracting explicit semantic information from a video stream; i.e. object detection, object tracking and information extraction; has been extensively investigated. However, little work has been devoted on the problem of discovering global or implicit...
Conference Paper
A fundamental issue in understanding the biological cellular behavior is based on discovering the interactions between genes, which is known as the gene regulatory network. This paper proposes a novel method to model large-scale gene regulatory networks from time series gene expression data. In the first step, a novel Gene Ontology (GO)-based clust...
Article
To print a colour image on a fabric, colour image clustering is an important step to reduce the number of colours and separate the coloured pattern. Therefore, the performance of colour-clustering algorithm can strongly affect the quality of printing process. Fuzzy c-mean (FCM) clustering is a known clustering algorithm for colour image quantizatio...
Article
Vulnerabilities in common security components such as firewalls are inevitable. Intrusion Detection Systems (IDS) are used as another wall to protect computer systems and to identify corresponding vulnerabilities. In this paper, a novel framework based on data mining techniques is proposed for designing an IDS. In this framework, the classification...
Article
In this paper, we propose a novel and effective algorithm for tracking soccer players in goal scenes, by eliminating fast camera motions effect through the correspondence between line marks in soccer field model and image sequences. The proposed algorithm comprises four steps. At the first step, we introduce an automatic grass field extraction algo...
Article
Steganography is the art and science of hiding secret data to provide a safe communication between two parties and it is a prominent branch in the information hiding research area. This paper presents a new steganographic method based on predictive coding and embeds secret message in quantized error values via Quantization Index Modulation (QIM). T...
Article
In this paper, we proposed a new clustering method where each cluster is created based on its characteristic that we call texture. Extraction of the texture relies on measuring the similarity of neighboring patterns. Our proposed clustering algorithm consists of two stages. In the first stage, sub-clusters are created based on the similarity of the...