Earticle

In recent years, with the development of compressive sensing (CS) theory, it has been widely applied to each field including image fusion, and obtained better fusion effect. And CS can reduce dimensions and the amount of data characteristics as well as the large amount and high computation complex. Therefore, this paper proposes a novel medical image fusion method based on compressive sensing theory in non-subsampled contourlet transform (NSCT) domain. First, NSCT transform is applied to the source images, and the coefficients in low frequency subband are fused by mean rules. For high frequency subband, CS is applied and the coefficients are fused by neighborhood-energy-MAX (NE-MAX) rule, then inverse CS is used to get fused coefficients. Finally, inverse NSCT is applied to get the reconstructed image. The experimental results show that the fusion algorithm proposed in this paper is superior to fusion method based on WT-MAX and CS-MAX、CS-MEAN.

The paper has investigated the meta-data for annotating the video information, and put up with the framework architecture of a semantic meta-data describing video information based on ontology. Firstly a domain ontology library is constructed, the video information is described by the ontology, and the video library system with architecture is developed. The paper builds a semantic ontology library for Beijing Opera, and gets the conclusion that it is feasible to describing video information with semantic meta-data based on ontology.

The performance of credit scoring models is determined by the used features. The relevant features for credit scoring usually are determined unsystematic and dominate by arbitrary trial. This paper presents a comparative study of four feature selection methods, which use data mining approach in reducing the feature space. The final results show that among the four feature selection methods, the Gini Index and Information Gain algorithms perform better than others with the classification accuracy of 75.46% and 75.44% respectively.

Build cutting resistance measurement system based on single-chip microcomputer and high speed camera in order to analysis cutting destruction of solid ice-snow. The experimental model has been built according to the rotating orthogonal experiment methods when tests, take the cutting resistance as objective function, A significant factor in the depth of cutting, rake angle as well as tool installation angle as influencing factors, and analysis found that the relationship between the primary and secondary factors were setting angle>density>cutting depth. By comparing theoretical models and experimental models found both the maximum error is about 8.7%, the minimum error is about 12.8%, the median error of about 12%, which verified the availability of theoretical model. Affecting the snow removal efficiency of parameters were carried out the optimal tests study, and found that the best setting angle of cutting tool is from 46°to 52°on the condition that the temperature is lower than -15°C, the thickness of snow and ice is less than 35mm, and rake angle -10°~ -9°.Optimal combination of parameters meet not only the job requirements but also provide a valuable reference for the optimum design of snow removal tool structure and process parameters, and the overall design of a prototype in the future.

The face recognition task involves extraction of unique features from the human face. Manifold learning methods are proposed to project the original data into a lower dimensional feature space by preserving the local neighborhood structure. PCA and LDA are used for the feature extraction and the resultant feature vectors are fused with the different fusion techniques and the proposed method yields good recognition rate with PCA Fusion of PCA and LDA features and those are classified with neural network. In general the size of the face database is too high and it needs more memory and needs more time for training so that to improve time and space complexities there is a need for dimensionality reduction. The extracted features are classified with Neural Network to improve the recognition rate.

This paper constructs the knowledge transfer simulation mode based on small world network model. It uses the average path length and clustering coefficient to stand for AC frequency and aggregation degree among knowledge network nodes and studies the node’s ability to release and absorb as well as the knowledge transfer effect by trust degree. Simulation results show that improving the AC frequency and aggregation degree of nodes, enhancing node’s ability to transfer knowledge can ensure transfer frequency in organization reaching a high level.

Members in electronic commerce often have no information about each other, thus, trust is one of the most important aspects in online transaction. In order to accurately display the buyers’ trustworthiness. This paper constructs a mean model to compute reputation of buyers. The model is based on two aspects, successful trading and returned goods. We introduce two adaptive trust factors, transaction value factor and return rate factor, to improve the feedback rates of sellers. Contrasting reputation based solely on feedback from sellers, the reputation of buyers is eventually represented by a three-dimensional array in our model. Simulation experiments evaluated different buyers and a buyer’s different period dynamic reputation. Compared with existing rating models, this model is more effective and can increase the ability to resist fraud.

This paper proposes a novel scheme for securing and authenticating digital data. Although, the experiments were applied to digital images, it is valid for sound, and video without any modifications of the core scheme. The developed scheme could be applied to the encoded data at the sender and the receiver sides, apart from the compression and decompression processes. The watermark information is generated automatically from the input data and hidden into the Least Significant Bits (LSBs) of specified pixels according to innovative criteria. Experimentally, the received data is authenticated if it is 100% match the sent data, exposing both geometric and signal processing attacks. The hidden watermark is completely invisible, achieving Peak Signal to Noise Ratio (PSNR) more than 67 decibels (dB) for grayscale images that vary in size and content. The developed scheme is evaluated by verifying images of the testing set before and after tampering, showing comparable performance. A comprehensive analysis for the experimental results revealed performance and efficiency of the proposed scheme is presented. An extensive review of previous work concerning image authentication approaches is presented, showing the need for robust and powerful authentication scheme.

Both the rapid growth of global economy and the rapid improvement in living conditions have witnessed an explosive requirement of foods that are rich of vitamin. Juice beverage, as one kind of these foods, is very popular among people of all ages due to its higher portability and easier preservability. However, there are many issues, such as a very low device utilization rate, a very long design period and an inflexible manufacturing process, by using the traditional juicing and filling production line. The introduction of virtual reality simulation technologies can not only detect bottlenecks of the production line during design period, but also process fruits according to seasons for improving the total efficiency of one single production line. In this paper, we combine customizable configuration and interactive control to design and implement a simulation system of juicing and filling production line based on the traditional client-server architecture. Practical applications show that the developed simulation system can not only shorten effectively the design period but also improve greatly the design quality.

This research is to investigate a structural model of an integrated information system for health promoting behaviors after radical prostectomy. The subjects of this paper were 142 patients who had visited a general hospital located in Metropolitan area. The pairwise t-test was done to compare the before and after application effect of health practice rate after radical prostatectomy. The results of this study are as follows. First, the response rate(29.6%) of experimental group was statistically significantly higher than the response rate(26.8%) in the 50 to 59 years old(χ²=10.27, p<.05). Second, for soybean intake, positive change of health behaviors behaviors diminished the progression rate of prostate cancer(31.24±0.35, t=-0.57, p=.000). Third, this paper found that the health promoting behaviors in prostate cancer patients was increased by 69.2-73.6% compared with the previous status. This study will contribute to reducing prostate cancer patients from prostatitis to prostate cancer in the future.

Disasters occur and are broadcasted by social media in our daily life. Retrieving the user-oriented relevant events in an aggregated results list through meta-search engine is a useful application. The definition and detection of events has been a hot topic in the study of natural language processing. Then gathering the disaster event information via different search interface for an event chain releasing is important. This paper designed an event-triggered model for events detection. As the emergency requirement of government as well as individual activities, disaster events detection is discussed as an example in this paper. A meta-search engine interface model is constructed with java based on the open source project Carrot2. And a H-T-E (Hazard, Trigger and Event) query expansion approach is applied in our event triggered model. The experimental results show that the proposed method can bring about increased accuracy and extra clues not supplied by commercial search engines. The solution we applied can be useful for other information supervision tasks and some issues like expecting the unexpected crisis etc. also triggered by the project for further study.

Quality is a major challenge throughout software projects, it is important to analyze and assess quality of a software project. Regular quality measurement is key when it comes to running a successful project. This paper presents a quantitative quality evaluation model of software project for Web application. This model uses Analytical Hierarchy Process (AHP), Fuzzy Mathematical theory and expert method to constrain quality metrics with respect. This model provides the weights of quality characteristics and sub-characteristics using analytic hierarchical process technique, which is on the basis of summering and analysis of historical test data for Web application. The experimental results indicate that this approach is effective for software project quality evaluation and can get good estimates.

The advent of advanced user interface devices has raised the interest of industry and academia in finding new modes of Human-Computer Interaction. Advanced interfaces employ gesture recognition, as well as motion and voice capturing to enable humans to interact naturally with interactive environments without utilizing any of the traditional devices like mice, joysticks or keyboards. Many approaches have been developed using a large variety of sensors to capture human interaction information and then provide further processing and recognition of the acquired information. However, the majority of these approaches usually focus on the actual implementation of the various stages that comprise an advanced interaction environment. Thus, the need for defining common data formats for improving integration and reutilization of these solutions are typically not addressed. On the other hand, this study aims at surveying existing research on integrating devices into interactive environments, at different interoperability levels and in data formats, identifying techniques and patterns of conveying information from the real world to the virtual world, in order to synthesize results, organize applicable documents by similarities and identify future research needs.

Visual analysis has potential to be used for recognition, and it is one of the hottest but most difficult subjects in computer vision. In order to identify pedestrian movement in an Intelligent Security Monitoring System, the video activity in the prospect is represented by a series of spatio-temporal interest points. Since human posture has the characteristics of uncertainty and illegibility, the clustering centers of each class are computed by fuzzy clustering techniques. We presented a pedestrian classification method based on improved support vector machines in order to solve non-rigid objects that are difficult to identify in intelligent monitoring systems. The Support Vector Machine technology and the decision tree have combined into one multi-class classifier so as to solve multi-class classification problems. Then a full-SVM (Support Vector Machine) decision tree is constructed based on the conventional decision tree. At last, the method is evaluated on the KTH action dataset and receives a comparatively high correct recognition rate.

The objective of transportation network equilibrium analysis is to know about the distribution of traffic flow on network, for ensuring network state and providing theoretical reference for improvement design of transportation network. The paper analyzes route choice behavior based on travel time reliability under uncertainty, and establishes road choice model based on travel time reliability. The paper proposes several travel time definitions. And the road choice model under uncertainty is applied to equilibrium distribution of transportation network to analyze the distribution of network flow of travel time reliability under different situations. And the paper analyzes the reality that several traffic modes exist on network, and takes accounts of the situations that traffic modes influence each other and don’t influence each other on network. The paper analyzes route choice behavior of traveler under information condition. Combined with the factors of mixed traffic and reliability route choice behavior, the paper researches the traffic equilibrium partitioning models under two distributing criterions.

The purpose of this paper are to analyze and design of knowledge management systems on Komisi Akreditasi Rumah Sakit (KARS) Indonesia - Indonesian Commission on Accreditation of Hospitals, especially for discussion forums and Content Management System, it will help to facilitate hospital accreditation in terms of accessing the necessary information and provide communication and learning for all parties involved in the accreditation. Analysis methodology that we use is the analysis and design of information systems, interviews, and literature review. The results achieved, the modules that we have are Discussion Forum and Content Management Systems (CMS). Forum is a means of media for discussion between KARS, hospital, surveyor and seminar speaker. While CMS allows KARS to provide learning materials and information about accreditation that can be downloaded by registered hospitals. In conclusion, expected by the knowledge management system can help KARS in managing information related to the accreditation of hospitals.

Modern Information technology has greatly improved the efficiency of knowledge exchange across time and distance. However, the corporations still face many difficulties to share knowledge because of many social factors such as corporate social responsibility. In this article, a frame model is established to analyze the relationship between corporation social responsibility and knowledge sharing by middle variables leadership ability and reputation. And then, Based on social network, an evolution model of knowledge sharing is constructed to describe how the corporate social responsibility affects the knowledge sharing among corporations. The simulation results firstly confirm that there is a significant positive correlation between them. Secondly, the taking CSR actively increases the leadership ability and reputation and improve knowledge sharing. But leadership ability expands its impact by increasing social network density, while reputation expands its impact by spillover effect. This provides a new perspective to study the non-technical factors which lead to dilemma of knowledge sharing. Surely, it is also helpful for us to deeply understand how the corporate social responsibility impacts the corporate performance.

In this paper, the classification of animal facial characteristics is examined for the purpose of creating anthropomorphic animal characters, followed by a description of the design process involved in creating a morphing program that can be used to generate various anthropomorphic animal characters. In order to establish an animal archetype database (DB) and an anthropomorphic animal character DB, we systematically analyze the facial characteristics of mammal characters. Subsequently, a technique is proposed with which to express component-based animal faces using a B-Spline curve in the morphing system. This program offers an intuitive user interface with a prototype generated upon the definition of necessary menus that allow system users to select optimally anthropomorphized images from an input source-target image morphing process.

Although standard DiffServ (Differentiated services) framework allows simultaneous transmission of video and data streams, it can not distinguish the importance of different video frames. Consequently, more received video packets may not lead to better decoded video quality. In this paper a novel inner-priority based video streaming scheme and its corresponding IP policy are proposed. Using IP policy, the importance of video packet could be recognized and corresponding RED queue will be assigned. Comprehensive evaluation results show that the average PSNR of IP policy is much higher than that of traditional policy in most cases and RED parameters settings influence the performance of IP policy. The best choice of RED parameters settings depends on many factors, including available bandwidth of the video stream, coding structure of video sequence, and the employed scheduling mode and dropping mode.

We present an approach for privacy preservation in cloud computing environment in which we propose to use information entropy and rough set theory with the goal of personalized privacy protection in cloud users during service selection based on trust. In the approach, cloud server is as an active service provider to personalize the cloud user privacy, cloud user selects the service which is quantified by trust for multi-tenant personalized privacy protection needs. We descript user privacy information by the service and trust attributes, design a multi-attribute service quantization algorithm using information entropy and rough set theory for the quality of service. The approach can achieve the purpose of quantitative service and dynamically adjust the supply relationship service based on multi-attribute, analysis shows which can effectively protect user privacy and personalization by quantifying the cloud service.

This paper presents a novel circular effect generation approach. In general, pixel location in an image can be represented with i and j coordinate. To manipulate (i,j) coordinate system easier, we transform (i,j) coordinate system into (ρ,θ) coordinate system. Two parameters are used for generating circular effect: R and T. After applying selected R and T in (ρ,θ) coordinate system, (ρ2,θ2) are obtained. Finally, (ρ2,θ2) signals are inverse-transformed into (i,j) coordinate system and (i2,j2) is obtained. Experimental results introduce performance comparison.

For the problem that general characteristic impedance measurement instrument is difficult to measure the size of clock jitter, this paper proposes a measurement algorithm to achieve clock jitter based on sequential equivalent sampling. First, according to the relationship between the equivalent sampling signal and the average signal, we count up the probability density distribution of the clock jitter and use the Tailfit jitter separation techniques to decompose total jitter (TJ) into deterministic jitter (DJ) and random jitter (RJ), then count up the peak-to-peak values of periodic jitter (PJ), cycle-to-cycle jitter (CCJ) and time interval error (TIE) to achieve a precise measurement of clock jitter. Finally we use the test results of jitter analysis software of Tektronix’s TDSJIT3 as a reference to verify the effectiveness of the algorithm. This algorithm provides quantitative analysis means for the stability of characteristic impedance measurement instrument, and provides the reference basis for self-calibration circuit design of subsequent hardware.

Orthogonal signal correction (OSC)and partial least squares(PLS)were used during the pretreatment of straw to reduce environmental noise and prediction models were established for near-infrared detection of straw cellulose. Tests were run with soybean stalk as the object of study. Test results indicated that compared to a model established using a traditional denoising method, the determination coefficients for calibration set models established by second derivative+smoothing and OSC were 0.9318595 and 0.9328905 respectively while the root mean square error for calibration (RMSEC) were 0.6762902 and 0.6696454. For an OSC-PLS regression model with a factor of 8, the relative standard deviation of a prediction model was less than 5%. In the OSC denoising process, the root mean square error fluctuated with the increasing number of PLS factors. Compared to the second derivative-smoothing denoising, OSC-PLS denoising removed the non-correlated variation from spectra and improved interpretation ability of variation while the analysis and convergence were expedited. It was therefore concluded that OSC-PLS denoising could be used to realize the rapid and accurate near-infrared detection of straw cellulose.

This paper presents CSGM2, a text preprocessing technique for compression purposes. It converts the original text into a word net (graph representation) and can retain the detailed contextual information such as word proximity. Specific directed graph is proposed to model this word net where words are stored in vertices and edges represent word transitions. The word net is fully capable of holding the natural word order in the original text and hence can be used directly for encoding purposes.

This paper considers the fuzzy modeling and robust fuzzy variable control for quadrotor in uncertain environment. The nonlinear system of quadrotor is firstly analyzed in this paper by utilizing laws of motion and force. Then, a Takagi-Sugeno (T-S) fuzzy model is achieved to approximate the system. LMI method is used to acquire an improved sliding surface. On this basis, a fuzzy variable structure controller is designed to force the system state trajectory toward sliding surface and maintain on it then. The controller ensures the resulting closed-loop quadrotor system is asymptotically stable. Finally, a simulation is shown to verify the effectiveness of the proposed algorithm.

This paper puts forward the methods on MOOCs knowledge organization and discovery, adopts the method of Formal Concept Analysis, uses concept lattice as the support tool, and selects the data from “Coursera” to cluster the courses and mine the inner knowledge association among courses, so as to discover the connotative knowledge correlation among MOOCs on the same topic and the structure characteristics among courses. Finally, based on the supergraph of the MOOCs concept lattice, the paper puts forward the visualization navigation method of the courses and provides flexible guiding principles for the learners whose knowledge structure is unusual in various fields when choosing the courses, so as to promote the development and improvement of the MOOCs websites by the knowledge organization and knowledge discovery technology.

The e-book industry has been growing very fast for the last several years, and the e-book sales at Amazon.com already surpassed the printed books at the end of 2010. In the near future, the printed textbooks in K-12 and in higher education will be replaced by the e-textbooks that support multimedia and user interaction. Furthermore, the future e-textbooks will support user interactive quizzes and instant feedbacks on the quizzes for students’ better learning and lecturers’ better teaching. Quizzes and feedback in classes are very common and helpful to check students’ understanding and to correct their misunderstanding. However, the current e-textbooks support only limited capabilities for quizzes and feedback. For example, they do not support interactive quizzes, instant feedbacks, statistics for students’ understanding, or methods for lecturers to make their quizzes in the e-textbooks. In this paper, we propose an e-textbook system that supports interactive quizzes, instant feedback, and quiz-maker, which allows lecturers to make their own quizzes at schools after deploying e-textbooks. Using this e-textbook system, lecturers can make quizzes at any pages in the e-textbooks, and students are able to check their understanding efficiently with instant feedback while reading the books. We also introduce our prototype system, which is implemented utilizing HTML5. The prototype system has advantages such as device independence, ease to implement and low cost.

Sound Source Localization (SSL) has a wide application in speech separation, recognition and enhancement. Binaural sound source localization based on human spatial hearing mechanism is an important research field of SSL. The recent binaural SSL research is focused on the system robust against noise and reverberation. In order to improve the localization performance in degraded environment, this paper proposes an algorithm to adaptively select the ‘good’ sub-bands to compute the binaural localization cues. Firstly, sub-band Signal-Noise Ratio (SNR) is estimated based on the auto-correlation matrix of binaural sound signals. Then, Inter-aural Time Difference (ITD) is computed by adaptively selecting the sub-bands which have the high SNR. Since the ITD is calculated through the sub-bands which are less affected by the noise, the sound source azimuth is estimated more accurate. The simulation results show that compared to the conventional binaural SSL algorithm, the localization accuracy of the proposed algorithm has been improved significantly.

In view of the fact that the propagation path topology cannot effectively deal with complex social network consists of hundreds of millions of users. More researchers choose to use machine learning methods to complete retweet prediction. Those use the classification method to judge whether a message will be retweeted or not. This paper argues that retweet prediction should be regression analysis problem, not just the classification problem. Through collecting user characteristics on Twitter and selecting some features which have an important impact on the retweet behavior, a Prediction algorithm Based on the Logistic Regression for users Retweet Times in social network was proposed. Experiment results based on the actual data set show the regression analysis predicting model has a good predicting accuracy in dealing with retweet predicting, the proposed method is effectiveness.

Along with the growth of the mobile game market, various and significant changes are taking place in terms of the game development environment, open policy, and genre. As one such change, “mobile-linked online games,” i.e., mobile games linked with online PC games, have appeared. The term linkage indicates that games, which were once independent of each other, have become connected through the linking of game data across mobile, console, PC, and Web platforms; this goes beyond the concept of a cross-platform, in which the same games are simply built on different platforms. A mobile platform developed for linking is obviously central to mobile-linked online games. Many linked online games are being produced on the mobile platform. However, in the case of online PC games, whose contents were originally developed for the PC platform, such as MMORPG (Massively Multiplayer Online Role-Playing Games), the method and purpose of their linkage differ. This paper systematically classifies mobile-linked online games into three types: mid-core, hardcore, and casual games. In addition, this paper reveals the characteristics and purpose of the online linking method for mobile-linked MMORPGs (a MMOPRG is a typical hardcore type game). Furthermore, the mobile linkage model for MMORPGs is defined by the PVE and PVP contents.