List of Articles
-
Open Access Article
1 - Blind Video Steganalysis by Semi-Supervised Approach for Motion Vectors Based Steganography Algorithms
Supervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, MoreSupervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, we cannot be sure about the effectiveness of steganalyzer in identifying non-accessible video steganography algorithms. On the other hand, using offline classification methods in the blind video steganalysis causes the learning process be time consuming and the system cannot be updated online. To solve this problem, we propose a new method for the blind video steganalysis by semi-supervised learning approach. In the proposed method, by eliminating the limitation of labeled training dataset, the classifier performance is improved for video steganography algorithms with non-accessible code. It is also proved that the proposed method, compared to common classification methods for the blind video steganalysis, has less time complexity and it is an optimal online technique. The simulation results on the standard database show that in addition to the above advantages, this method has appropriate accuracy and is comparable to common methods. Manuscript profile -
Open Access Article
2 - Blind Video Steganalysis by Semi-Supervised Approach for Motion Vectors Based Steganography Algorithms
J. Mortazavi Mehrizi M. Khademi H. Sadoghi YazdiSupervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, MoreSupervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, we cannot be sure about the effectiveness of steganalyzer in identifying non-accessible video steganography algorithms. On the other hand, using offline classification methods in the blind video steganalysis causes the learning process be time consuming and the system cannot be updated online. To solve this problem, we propose a new method for the blind video steganalysis by semi-supervised learning approach. In the proposed method, by eliminating the limitation of labeled training dataset, the classifier performance is improved for video steganography algorithms with non-accessible code. It is also proved that the proposed method, compared to common classification methods for the blind video steganalysis, has less time complexity and it is an optimal online technique. The simulation results on the standard database show that in addition to the above advantages, this method has appropriate accuracy and is comparable to common methods. Manuscript profile -
Open Access Article
3 - Blind Video Steganalysis by Semi-Supervised Approach for Motion Vectors Based Steganography Algorithms
J. Mortazavi Mehrizi M. Khademi H. Sadoghi YazdiSupervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, MoreSupervised learning algorithms are widely used in blind video steganalysis and the cost of generating labeled data in them is high. That is why only a limited number of steganography algorithms with accessible code can be used for the training the classifier. Therefore, we cannot be sure about the effectiveness of steganalyzer in identifying non-accessible video steganography algorithms. On the other hand, using offline classification methods in the blind video steganalysis causes the learning process be time consuming and the system cannot be updated online. To solve this problem, we propose a new method for the blind video steganalysis by semi-supervised learning approach. In the proposed method, by eliminating the limitation of labeled training dataset, the classifier performance is improved for video steganography algorithms with non-accessible code. It is also proved that the proposed method, compared to common classification methods for the blind video steganalysis, has less time complexity and it is an optimal online technique. The simulation results on the standard database show that in addition to the above advantages, this method has appropriate accuracy and is comparable to common methods. Manuscript profile -
Open Access Article
4 - Content Based Image Retrieval by the Fusion of Short Term Learning Methods
B. Bagheri M. Pourmahyabadi H. Nezamabadi-pourContent based image retrieval (CBIR) contains a set of techniques to process the visual features of a query image, in order to retrieve images semantically similar to it, in a database. To improve the performance of image retrieval systems, relevance feedback tool can b MoreContent based image retrieval (CBIR) contains a set of techniques to process the visual features of a query image, in order to retrieve images semantically similar to it, in a database. To improve the performance of image retrieval systems, relevance feedback tool can be used. In this research, to increase the effectiveness of the image retrieval systems, the fusion of two (multiple) short term learning methods based on relevance feedback is proposed. In the proposed method, fusion is performed in three levels: fusion in ranks, fusion in retrieved images, and fusion in similarities. To evaluate the performance of the proposed method, a CBIR system with 10000 images of 82 different semantic groups is employed. The experimental results confirm the superior of suggested method in terms of retrieval precision. Manuscript profile -
Open Access Article
5 - Content Based Image Retrieval by the Fusion of Short Term Learning Methods
B. Bagheri M. Pourmahyabadi H. Nezamabadi-pourContent based image retrieval (CBIR) contains a set of techniques to process the visual features of a query image, in order to retrieve images semantically similar to it, in a database. To improve the performance of image retrieval systems, relevance feedback tool can b MoreContent based image retrieval (CBIR) contains a set of techniques to process the visual features of a query image, in order to retrieve images semantically similar to it, in a database. To improve the performance of image retrieval systems, relevance feedback tool can be used. In this research, to increase the effectiveness of the image retrieval systems, the fusion of two (multiple) short term learning methods based on relevance feedback is proposed. In the proposed method, fusion is performed in three levels: fusion in ranks, fusion in retrieved images, and fusion in similarities. To evaluate the performance of the proposed method, a CBIR system with 10000 images of 82 different semantic groups is employed. The experimental results confirm the superior of suggested method in terms of retrieval precision. Manuscript profile -
Open Access Article
6 - Semi-Supervised Learning Based on Extreme Learning
A. Mehrizi H. Sadoghi Yazdi S. J. Seyyed Mahdavi ChabokSemi-supervised learning with growing self-organizing map (GSOM) is used in many applications, such as clustering. The main challenges in the Semi-supervised GSOM are calculating parameters such as shape and structure of clustering layer, activation level, and weights o MoreSemi-supervised learning with growing self-organizing map (GSOM) is used in many applications, such as clustering. The main challenges in the Semi-supervised GSOM are calculating parameters such as shape and structure of clustering layer, activation level, and weights of classifier layer. Current approaches use initiative methods with a local look have trying to determine these parameters; which its effect, the results of these algorithms is highly dependent on the conditions. This paper studies a semi-supervised learning method based on GSOM and extreme learning for the first time. The proposed method, without the direct calculation of the GSOM parameters and using the extreme learning determines label of each data. Error resulted from the feedback system is used to optimize extreme learning and GSOM. In this paper, in addition to investigating the convergence analysis of the proposed method, sequential extreme learning is also provided for semi-supervised GSOM. Experiments conducted on online and partially labeled data show that the proposed method has a relative advantage in terms of accuracy on semi-supervised GSOM. Manuscript profile -
Open Access Article
7 - Learners Grouping in Adaptive Learning Systems Using Fuzzy Grafting Clustering
M. S. Rezaei Gh. A. MontazerQuality of adaptive and collaborative learning systems is related to appropriate specifying learners and accuracy of separation learners in homogenous and heterogeneous groups. In the proposed method for learners grouping, researchers effort to improving basic clusterin MoreQuality of adaptive and collaborative learning systems is related to appropriate specifying learners and accuracy of separation learners in homogenous and heterogeneous groups. In the proposed method for learners grouping, researchers effort to improving basic clustering methods by combination of them and improving methods. This work makes the complexity of grouping methods increased and quality of result’s groups decreased. In this paper, new method for selection appropriate clusters based on fuzzy theory is proposed. In this method, each cluster is defined as a fuzzy set and the corresponding clusters are determined. So the best cluster is selected among each corresponding clusters. The results of an empirical evaluation of the proposed method based on two criteria: “Davies-Bouldin” and “Purity and Gathering” indicate that this method has better performance than other clustering methods such as FCM, K-means, hybrid clustering method (HCM), evolutionary fuzzy clustering (EFC) and ART neural network. Manuscript profile -
Open Access Article
8 - Evaluation of Fuzzy-Vault-based Key Agreement Schemes in Wireless Body Area Networks Using the Fuzzy Analytical Hierarchy Process
M. Ebrahimi H. R. Ahmadi M. Abbasnejad AraWireless body area networks (WBAN) may be deployed on each person’s body for pervasive and real time health monitoring. As WBANs deal with personal health data, securing the data during communication is essential. Therefore, enabling secure communication in this area ha MoreWireless body area networks (WBAN) may be deployed on each person’s body for pervasive and real time health monitoring. As WBANs deal with personal health data, securing the data during communication is essential. Therefore, enabling secure communication in this area has been considered as an important challenge. Due to the WBAN characteristics and constraints caused by the small size of the nodes, selection of the best key agreement scheme is very important. This paper intends to evaluate different key agreement schemes in WBANs and find the best one. To achieve this goal, three schemes from existing research named OPFKA, PSKA and ECG-IJS are considered and a fuzzy analytical hierarchy process (FAHP) method is employed to find the best scheme. Manuscript profile -
Open Access Article
9 - A Distance-Based Method for Inconsistency Resolution of Models
R. Gorgan Mohammadi Ahmad Abdollahzadeh BarforoushModel driven approach to software engineering has been taken into consideration due to its impact on reducing complexities and improving the productivity in software development. Inconsistencies are considered as an important challenge in applying models. An inconsisten MoreModel driven approach to software engineering has been taken into consideration due to its impact on reducing complexities and improving the productivity in software development. Inconsistencies are considered as an important challenge in applying models. An inconsistency is occurred due to an undesired structural pattern in a model. The main drawback of current approaches to inconsistency resolution is not considering the difference between the repair and the spoiled model. This work presents a distance-based method for finding closest repair for the spoiled model. For this aim, models and metamodels are represented using directed graphs and graph transformation rules are employed for inconsistency resolution. A distance metric is defined based on the amount of changes in the graph corresponding to the model. Application of the proposed method to a set of BPMN models shows the improvement of the results. Manuscript profile -
Open Access Article
10 - Fault Detection by Integrating Canonical Variate Analysis and Independent Component Analysis Based on Local Outlier Factor
E. Tavasolipour M. T. Hamidi Beheshti A. RamezaniIn this paper a novel process monitoring scheme is proposed because of the importance of fault detection and identification in industrial processes. In this method, process dynamic and effect of outliers are considered concurrently. First, the proposed approach uses CVA MoreIn this paper a novel process monitoring scheme is proposed because of the importance of fault detection and identification in industrial processes. In this method, process dynamic and effect of outliers are considered concurrently. First, the proposed approach uses CVA method to implement the process dynamic. Then ICA method is performed for dimension reduction of data. The outliers elimination and control limit calculation are based on the Local Outlier Factor algorithm. This algorithm doesn’t consider a special distribution for process variables, thus conforming to data in real industrial processes. The proposed method is applied to fault detection in the Tennessee Eastman process. Results clearly indicate better performance of the proposed scheme compared to the alternative methods. Manuscript profile -
Open Access Article
11 - Separating Bichromatic Point Sets by Right Triangles
Z. Moslehi A. BagheriSeparating colored point sets is an interesting problem in computational geometry with application in machine learning and pattern recognition. In this problem, we are given a geometric shape C and two point sets P and Q of total size n as red and blue points, respectiv MoreSeparating colored point sets is an interesting problem in computational geometry with application in machine learning and pattern recognition. In this problem, we are given a geometric shape C and two point sets P and Q of total size n as red and blue points, respectively. Now, we must separate red and blue points by this shape such that all the blue points lie inside it and all the red points lie outside it. In the previous work, we have some algorithms for rectangle and wedge separability but we do not have any algorithm for separating by a triangle and separating by a triangle with a fixed angle such as right triangle. In this paper, we present an efficient algorithm for right triangle seprability. In this algorithm, we use sweep line technique and introduce some events and process them. So, we can report all separating right triangles in O(nlog n) time. Manuscript profile -
Open Access Article
12 - An Efficient Bread First Search Algorithm on CPU and GPU
P. Keshavarzi H. Deldari S. AbrishamiGraphs are powerful data representations used in enormous computational domains. In graph-based applications, a systematic exploration of graph such as a breath first search often is a fundamental component in the processing of the vast data sets. In this paper we prese MoreGraphs are powerful data representations used in enormous computational domains. In graph-based applications, a systematic exploration of graph such as a breath first search often is a fundamental component in the processing of the vast data sets. In this paper we presented a hybrid method that in each level of processing of graph chooses the best implementation of algorithms implemented on CPU or GPU, while avoid poor performance on low and high degree graphs. Our method shows improved performance over the current state-of-the-art implementation and our results proves it. Manuscript profile