جستجوی مقالات مرتبط با کلیدواژه « همبستگی » در نشریات گروه « برق »

This paper presents a reversible data hiding method in encrypted image that employs correlation of neighboring pixels in the image. In the proposed method, original image may be encrypted by desire encryption algorithm. More significant bits of the pixels in the image are exploited to vacate room for embedding data bits. In the approach, image is divided into separated blocks and most central pixel of each block is considered as reference one. The prediction error between the intensity of other pixels and reference one is calculated and denoted local prediction. This error is analyzed determining a feature of block embedding capacity. Calculated features for all blocks are compressed employing arithmetic coding and embedded in the image along with data bits. At the recipient, at first, compressed features are extracted, then they are uncompressed and used to lossless reconstruction of the original image and extraction of the data bits. Experimental results confirm that the proposed algorithm outperforms state of the art ones.

Today, the continuous changes in customer requirements are the main challenges faced by enterprises. Service-oriented architecture is considered as a practical solution to solve this problem for service-oriented enterprises. In the service-oriented architecture, selection and composition of services to quickly respond to complex customer requirements is available to service-oriented enterprises. Enterprises use ready-to-use and outsourced services to respond more quickly to the complex and changing needs of customers. One of the emerging technologies in this area is web services. By expanding the desire of enterprises to use web services, overtime web services providers increased. For this reason, Web services with the same functionality and different qualities were expanded. Therefore, the issue of choosing a web service with the best quality for enterprises is important. On the other hand, enterprises with only one web service cannot meet the complex requirements of customers; therefore, they need to composite multiple web services together. In addition, with the increase of web services with different functions, correlation, dependency and conflict between Web services also expand in their composition. But so far, there is no way to choose the best web services based on the quality of service(QoS) and also their composition does not violate the dependency, conflict and correlation between web services. In this paper, we try to make use of previous methods that consider dependency or conflict or correlation in simple modes of web services composition. We will improve all these methods in a comprehensive approach and support complex situations that may arise from the composition of web services and find the suitable composite web service by considering dependency, conflict, and correlation between Web services.

Most of the integrated components used in current CMOS integrated circuits (IC) technology are inevitably nonlinear. This issue complicates the matching between such elements, and affects strongly the performance of analog circuits. In data converters, the nonlinearity caused by employing nonlinear transistors lowers the overall resolution and may limit the number of bits to an unacceptable value. It can be demonstrated that the number of the stages that should be calibrated is equal to the difference between the maximum number of bits without calibration and the desired number of bits which should be calibrated. In this article, we modeled the main error sources of the gain factor in pipeline stages. A modified calibration technique is then applied to estimate and to calibrate the nonlinear gain factor of the primitive stages. The effectiveness of the proposed approach is verified through design and calibration of a 14-bit 65MS/s converter in 0.18µm standard CMOS technology.

The presence of missing values in the real world data is a very prevalent and inevitable problem. So, it’s necessary to fill up these missing values accurately, before they are used for knowledge discovery process. This paper proposes three novel methods to fill numeric missing values. All of the proposed methods apply regression models on subsets of data which there are strong correlations among them. These subsets are selected using forward selection based approaches. In the selection of the desired subsets, it is tried to maximize the correlation between missing attribute and other attributes. The correlation coefficient is used to measure the relationships between attributes. The priority of each missing attribute for imputation purpose is also considered in the proposed methods. The performance of proposed methods is evaluated on five real world datasets with different missing ratios. The efficiency of the proposed methods is compared with five different estimation methods, namely, the mean imputation, the k nearest neighbours imputation, a fuzzy c-means based imputation, a decision tree based imputation, and a regression based imputation algorithm, called “Incremental Attribute Regression Imputation” (IARI) method. Two well-known evaluation criteria, namely, Root Mean Squared Error (RMSE) and Coefficient of Determination (CoD) are used to compare the performance of proposed methods with other imputation methods. Experimental results show that the proposed methods perform better than other compared methods, even when the missing ratio is high.

Headways (i.e. the time period between the departure times of two consecutive transportation vehicles) is an important issue for urban railway companies. In this problem we are facing with two conflicting objectives, average passenger travel time and rate of carriage fullness. Until now different multi-response optimization procedure for solving this problem is studied but these approaches do not consider responses variance and covariance between them in optimization process. Therefore this research presents a modeling and solution approach based on discrete-event simulation and response surface methodology that not only puts average of response in satisfactory region but also try to minimize the variance of responses relative to noise variable and also consider covariance between them. In order to evaluate the performance of the proposed approach, Tehran metro line 4 has been assessed. The results show that the proposed approach is superior to existing techniques.

Video-based face recognition has received significant attention in recent years. There are some problems in video-based face recognition method such as pose, lighting and expression variations, occlusion, aging and low image resolutions. In this paper, first of all face in video frames is detected for background removing. Then each set of images is distributed on a nonlinear manifold and clustered using appropriate methods. Determination of the number of clusters and linear models is quite intelligent and with implementation of this method, the number of clusters depends on different movements of head. Two methods are proposed for clustering and linear models. The distance between nonlinear manifolds that is composed of the combination of two methods, are used for recognition. The first method uses the distance between the clusters which are considered as the linear space and the second method uses the distance between key frames for recognition. Finally the results are compared with other methods.