Sharif Digital Repository

Sport video shot classification is a basic step in the sport video processing. For many purposes such as event detection and summarization, shot classification is needed for content filtering. In this paper, we present a new method for soccer video shot classification. At first, in-field and out-of-field frames are separated. In in-field frames three features based on number of connected components and shirt color percent in vertical and horizontal strips are extracted. The features are all new and showed excellent discrimination in the feature space. These features are given to SVM for classifying long, medium and close-up shots. One of the advantages of our method is that, close-ups can be... 

This paper presents a new run-based algorithm for labeling connected components in a binary image. The algorithm removes assumption on all border pixels of image are background. Also it does not use merging operation for resolving label equivalences among provisional labels (sets), but it uses post processing stage. The post processing stage reduces complexity for resolving label equivalency. During the first scan, provisional labels are assigned to the connected components. After the first scan, the post processing is done to resolve label equivalency. The smallest provisional label among all provisional labels that are assigned to a connected component is considered as a representative... 

In this paper the Unsupervised Active Learning Method (UALM), a novel clustering method based on the Active Learning Method (ALM) is introduced. ALM is an adaptive recursive fuzzy learning algorithm inspired by some behavioral features of human brain functionality. UALM is a density-based clustering algorithm that relies on discovering densely connected components of data, where it can find clusters of arbitrary shapes. This approach is a noise-robust clustering method. The algorithm first blurs the data points as ink drop patterns, then summarizes the effects of all data points, and finally puts a threshold on the resulting pattern. It uses the connected-component algorithm for finding... 

Hierarchical reinforcement learning facilitates learning in large and complex domains by exploiting subtasks and creating hierarchical structures using these subtasks. Subtasks are usually defined through finding subgoals of the problem. Providing mechanisms for autonomous subgoal discovery and skill acquisition is a challenging issue in reinforcement learning. Among the proposed algorithms, a few of them are successful both in performance and also efficiency in terms of the running time of the algorithm. In this paper, we study four methods for subgoal discovery which are based on graph partitioning. The idea behind the methods proposed in this paper is that if we partition the transition... 

Hierarchical reinforcement learning (HRL) has had a vast range of applications in recent years. Preparing mechanisms for autonomous acquisition of skills has been a main topic of research in this area. While different methods have been proposed to achieve this goal, few methods have been shown to be successful both in performance and also efficiency in terms of time complexity of the algorithm. In this paper, a linear time algorithm is proposed to find subgoal states of the environment in early episodes of learning. Having subgoals available in early phases of a learning task, results in building skills that dramatically increase the convergence rate of the learning process. © 2009 Springer... 

This paper models the cyber-social system as a cyber-network of agents monitoring states of individuals in a social network. The state of each individual is represented by a social node, and the interactions among individuals are represented by a social link. In the cyber-network, each node represents an agent, and the links represent information sharing among agents. The agents make an observation of social states and perform distributed inference. In this direction, the contribution of this paper is threefold: First, a novel distributed inference protocol is proposed that makes no assumption on the rank of the underlying social system. This is significant as most protocols in the... 

In this paper we define critical graphs as minimal graphs that support a given set of rates for the index coding problem, and study them for both the one-shot and asymptotic setups. For the case of equal rates, we find the critical graph with minimum number of edges for both one-shot and asymptotic cases. For the general case of possibly distinct rates, we show that for one-shot and asymptotic linear index coding, as well as asymptotic non-linear index coding, each critical graph is a union of disjoint strongly connected subgraphs (USCS). On the other hand, we identify a non-USCS critical graph for a one-shot non-linear index coding problem. In addition, we show that the capacity region of... 

This paper presents a neural network-based method for automatic classification of magnetic resonance images (MRI) of brain under three categories of normal, lesion benign, and malignant. The proposed technique consists of six subsequent stages; namely, preprocessing, seeded region growing segmentation, connected component labeling (CCL), feature extraction, feature Dimension Reduction, and classification. In the preprocessing stage, the enhancement and restoration techniques are used to provide a more appropriate image for the subsequent automated stages. In the second stage, the seeded region growing segmentation is used for partitioning the image into meaningful regions. In the third... 

Nowadays, we are facing epidemic spreading in many different areas; examples are infection propagation, rumor spreading and computer viruses in computer networks. Finding a strategy to control and mitigate the spread of these epidemics is gaining much interest in recent researches. Due to limitation of immunization resources, it is important to establish a strategy for selecting nodes which has the most effect in mitigating epidemics. In this paper, we propose a new algorithm that minimizes the worst expected growth of an epidemic by reducing the size of the largest connected component of the underlying contact network. The proposed algorithm is applicable to any level of available resources... 

Face Analysis is an important task in image processing. Most of these tasks centralized on face recognition and detection. One of different ways for deceiving automatic face analysis systems is mark notation on the skin. On the other hand some applications attempts to eliminate defects of the face. Hence, in this paper we try to detect and remove skin marks on the face, whether they're natural or not. Our algorithm passes face image through appropriate filters to get mark candidates and then create a graph space using 8-point neighborhood relations of mark candidates image pixels. Then we compute probabilities of each mark candidate using four measures based on intensity of occurrence, shape... 

In this paper we investigated the structural properties of the ultra high voltage power transmission network of Iran. We modeled the power grid as a network with 105 nodes and 142 connection links. We found that the Iranian power grid displays a relatively moderate clustering coefficient-much larger than that of corresponding random networks-and small characteristics path length comparable to that of corresponding random networks; i.e. the power grid is a small-world network with exponential degree distribution. Global efficiency was considered as an indicator of grid's performance and the influence of random and intentional nodal failures on the efficiency was investigated. We also studied... 

Many real-world networks have a modular structure, and their component may undergo random errors and/or intentional attacks. More devastating situations may happen if the network components have a limited load capacity; the errors and attacks may lead to a cascading component removal process, and consequently, the network may lose its desired performance. In this brief, we investigate the tolerance of cascading errors and attacks in modular small-world networks. This brief studies the size of the largest connected component of the networks when cascading errors or attacks occur. The robustness of the network is tested as a function of both the intermodular connection and intramodular... 

Single-time-scale distributed estimation of dynamic systems via a network of sensors/estimators is addressed in this letter. In single-time-scale distributed estimation, the two fusion steps, consensus and measurement exchange, are implemented only once, in contrast to, e.g., a large number of consensus iterations at every step of the system dynamics. We particularly discuss the problem of failure in the sensor/estimator network and how to recover for distributed estimation by adding new sensor measurements from equivalent states. We separately discuss the recovery for two types of sensors, namely α and β sensors. We propose polynomial-order algorithms to find equivalent state nodes in graph... 

The hierarchical structure of real-world problems has resulted in a focus on hierarchical frameworks in the reinforcement learning paradigm. Preparing mechanisms for automatic discovery of macro-actions has mainly concentrated on subgoal discovery methods. Among the proposed algorithms, those based on graph partitioning have achieved precise results. However, few methods have been shown to be successful both in performance and also efficiency in terms of time complexity of the algorithm. In this paper, we present a SCC-based subgoal discovery algorithm; a graph theoretic approach for automatic detection of subgoals in linear time. Meanwhile a parameter tuning method is proposed to find the... 

Data stream classification is one of the most challenging areas in the machine learning. In this paper, we focus on three major challenges namely infinite length, concept-drift and concept-evolution. Infinite length causes the inability to store all instances. Concept-drift is the change in the underlying concept and occurs in almost every data stream. Concept-evolution, in fact, is the arrival of novel classes and is an undeniable phenomenon in most real world data streams. There are lots of researches about data stream classification, but most of them focus on the first two challenges and ignore the last one. In this paper, we propose new method based on ensembles whose classifiers use... 

In this paper, we study the problem of learning the structure of Bayesian networks from data, which takes a dataset and outputs a directed acyclic graph. This problem is known to be NP-hard. Almost most of the existing algorithms for structure learning can be classified into three categories: constraint-based, score-based, and hybrid methods. The K2 algorithm, as a score-based algorithm, takes a random order of variables as input and its efficiency is strongly dependent on this ordering. Incorrect order of variables can lead to learning an incorrect structure. Therefore, the main challenge of this algorithm is strongly dependency of output quality on the initial order of variables. The main...