Computer Science

Survey of Open Source Software for Building Cloud Computing Platforms

Computer Science. 2012, 39 (11): 1-7.

Abstract

PDF(694KB) ( 1176 )

RelatedCitation | Metrics

The emergence of open source software for cloud computing facilitates the building of cloud computing plat- forms, but meanwhile it challenges the choosing of appropriate ones from these pool of software. I}o figure out how to build a cloud computing platform,research on existing open source software for building cloud computing platforms is necessary. This paper investigated the development of these kinds of open source software and analyzed their architec- lures from the perspective of service models. Through comparison and analysis of representatives of them, effective ways were proposed for developers to choose appropriate pieces of software to build a specific cloud computing plat- form.

Rete Algorithm;Current Issues and Future Challenge

Computer Science. 2012, 39 (11): 8-12.

Abstract

PDF(535KB) ( 3294 )

RelatedCitation | Metrics

The production system is widely used in AI. But it's not practicable in large amount of data because of its bad matching efficiency. Rete algorithm,which speeds up the match efficiency by sharing condition elements and reserving temporary matching results, has become one of the most widely used reasoning algorithms for production systems. I}his paper gave a survey about the principles, state of art and facing problems of Rete algorithm. Many prominent improve- menu of Rete algorithmare were analyzed and compared. Finally, it concludes the challenges that Rete faces and the di- rection of future research.

Research on a Series of Problems in RBAC Model

Computer Science. 2012, 39 (11): 13-18.

Abstract

PDF(633KB) ( 1144 )

RelatedCitation | Metrics

RI3AC is characterized by distributed management and self-management as the basic model in RBAC field. Today, many research themes on RI3AC arc almost proceeded in the field of information science, and its management and logistics background are ignored. This paper uncovered a series of management problems and logic problems existed during the research process on RI3AC, including false ternary logic basis of authorized state, un-synchronization between right and responsibility of administrators, ill-defined meaning of right leakage, unclear resource of authority, and failure in decision-making during authorizing process etc. Then it analyzed the false ternary logic problem from a logic view with a three layer framework in detail, described the background, content and deriving relationships of other manage- ment problems from a management view, explained the meaning of right leakage and origin resource of authority, put forward the mechanism on the synchronization of right and responsibility and the corresponding audit system, and de- signed the decision support mode for the administrators when authorizing. These research can provide theoretical sup- port for the development and update of RI3AC model.

PS-PPF; VANETs Privacy Protecting Framework Based on Proxy Signatures Cluster

Computer Science. 2012, 39 (11): 19-23.

Abstract

PDF(464KB) ( 798 )

RelatedCitation | Metrics

VANETs are one of the most important Internet of things(IoT) applications in the intelligent transportation field,and arc the academic and industry research priorities in recent years. VANEI's have attractive prospects for devc lopment, but its application is strictly constrained by security and privacy, so the research about security and privacy of VANETs is becoming a hot spot, and there has been a large number of research results. Based on proxy signature clus- ter,we proposed PS-PPF;a VANETs privacy protecting framework. The framework not only can provide good identity and location privacy protecting for the vehicles, but also can be implemented and deployed well because of its practice based design and expandability. There are 4 levels, 7 key roles, and 4 important proxy signatures in PS-PPF. We also in- troduced some extensible explanations for the framework. Our proxy-signature-based framework is the first one for the practiccoricnted VANETs privacy protecting framework.

Modeling of CPS Simulation Based on the Integration of Heterogeneous Models

Computer Science. 2012, 39 (11): 24-28.

Abstract

PDF(446KB) ( 1152 )

RelatedCitation | Metrics

Single discrete system simulation or continuous system simulation can't meet the complex needs of CPS si- mulation. We divided the integrated model of CPS into three types of entities, built discrete event simulation model (UML model) based on event driven for the computation entity, continuous-time simulation model (Simufink model) based on timcdriven for the physical entity. We applied co-simulation to construct interactive entity, characterizing the interfaces and behaviors of computation entity and physical entity. This paper introduced the advantages of the collabo- rative modeling of UMI./Simulink, investigated and analyzed three different methods for model merging. At the model levcl,we designed the fusion of the CPS integrated model on the basis of unmanned aerial vehicle model.

DiffServ-based Network Coding Protocol in Heterogeneous Wireless Sensor Networks

Computer Science. 2012, 39 (11): 29-33.

Abstract

PDF(441KB) ( 678 )

RelatedCitation | Metrics

By combining multiple packets in a single broadcast transmission, network coding can greatly improve the ca- pacity of multi hop wireless networks. I}his paper proposed a diffscrv-based network coding protocol (DI3NC) for hetero- geneous wireless sensor networks. The main idea is to compute periodically the mean and standard deviation of resi-dual energy of nodes, together with a dynamic node classification. On this basis, the QoS reduirements are mapped to network coding metrics. For this purpose, we introduced an energy-aware forwarding mechanism for coded-packet. Meanwhile, a multi-priority packet scheduling strategy was designed to increase the opportunities for coding. Our simulation results demonstrate that DI3NC achieves a significant improvement in energy efficiency, as well as in reliability of data transmis- lion.

Research on P2P Traffic Control Based on Feedback Theory and Kalman Algorithm

Computer Science. 2012, 39 (11): 34-36.

Abstract

PDF(265KB) ( 1094 )

RelatedCitation | Metrics

With the increasing number of P2P users, P2P applications arc consuming most of the Internet bandwidth, otherwise they have a negative effect on traditional Internet applications. In order to solve thwas problem, a P2P traffic control method based on feedback principle and Kalman algorithm was presented. I3y means of feedback principle in con- trol theory and Kalman algorithm, the P2P traffic can be regulated and controlled properly. The experimental results show it can not only reduce the high P2P traffic, but also improve the low P2P traffic. It can make the peers to respond to the changes of network status rapidly and utilize the bandwidth sufficiently.

Link State Reasoning Based Optimized Routing Protocol for Wireless

Computer Science. 2012, 39 (11): 37-40.

Abstract

PDF(456KB) ( 1462 )

RelatedCitation | Metrics

Based on the analysis of the most important challenges in wireless Mesh networks, a link state reasoning based routing protocol was presented which is originated from optimizes link state routing protocol (LR-OLSR) with cross-layer design theory. hhe I_R-OI_SR protocol greatly optimizes routing performance by introducing the reasoning method to value the link quality depending on information such as the node load, the packet delivery ratio, and link avail bility. The proposed protocol makes use of link duality as the routing metric during route selection, thus it can achieve optimal routing and load balancing in wireless Mesh networks. The simulation results show that the proposed routing protocol improves the packet delivery ratio, reduces the end-to-end delay, and achieves load balance in the route selection process.

Incentive Mechanisms for Multicast Nodes Based on Second-price Auction Theory in P2P Network

Computer Science. 2012, 39 (11): 41-44.

Abstract

PDF(445KB) ( 691 )

RelatedCitation | Metrics

The reliability of P2P multicast performance can't be ensured when facing non-cooperation stategies and ma- licious behaviours of P2P nodes. The market model based framework of P2P multicast was proposed, which uses second- price auction for network resource allocation. The stategies and payoff matrixes of various multicast nodes were ana- lyzed,as well as the conditions of stategy adjustment. Then the incentive mechanisms were designed for multicast nodes. Theoretical analysis and simulation results demonstrate its effectiveness.

Self-adapting Data Gathering Scheme for Mobile Sink Node

Computer Science. 2012, 39 (11): 45-50.

Abstract

PDF(533KB) ( 760 )

RelatedCitation | Metrics

Mobile sink for wireless sensor network is an efficient way to prolong the network lifetime. We studied a sce- nario where the mobile sink node is located in a duadcopter which is in charge of data gathering from the wireless sensor network. After analyzing the properties of mobile sink node during flying, we gave the theoretical constraints for the mobile sink node in terms of mobile trace, velocity and height for the low power wireless communication. According to these constraints, we proposed the data amount based local optimal mobile trace control approach for the mobile sink node, so as to gather from the sensor network efficiently. Explicit simulations show that the proposed approach is feasi- ble, which provides a theoretical basis for such applications.

Botnet Propagation Model with Variable Infection Rate

Computer Science. 2012, 39 (11): 51-53.

Abstract

PDF(258KB) ( 751 )

RelatedCitation | Metrics

The threat of botnet on network security has been attached great importance by many security search ex- pens. Mathematical modeling is a kind of effective method to study botnet propagation. Existing botnet propagation models assume that infection rate is constant. However in fact, network congestion will be caused when a large number of botnet viruses rampant. Against this backdrop, a botnet propagation model with variable infection rate was proposed, which considers pre-immunity according to the real situation of Internet. Finally, the difference of botnet infection be- tween the model of this paper and existing models was depicted in matlab. Experimental results show that the botnet propagation model with variable infection rate is more accurate to reflect the practical propagation law of hots.

Epidemic Dynamics with Traffic Flow on Scale-free Networks

Computer Science. 2012, 39 (11): 54-57.

Abstract

PDF(338KB) ( 748 )

RelatedCitation | Metrics

Different from the classical epidemic models, the infectivity of a node is decided by its actual betweenness从 in the epidemic model based on traffi}flow other than degree k. Utilizing the mean-field theory,we proposed a modified SIS epidemic model based on traffi}flow. With this model,taking MIP route as an example,we re-studied the relation- ship between the spreading probability召，the traffic generation rate } and the epidemic threshold八，the stationary densi- ty of infected nodes p considering the scenarios in which the delivery capability of the nodes is bounded or unbounded.

Improvement and Formal Analysis of a Fair Multi-party Certified Mail Protocol

Computer Science. 2012, 39 (11): 58-61.

Abstract

PDF(382KB) ( 681 )

RelatedCitation | Metrics

Fair multi-party certified mail protocol is a value service to deliver important data over Internet with guaran- teed receipt for each successful delivery. Therefore, the protocol needs to be confidential, and meets non-repudiation, fairness,exclusion and timeliness. This paper pointed out potential security concern for a current protocol,and improved the protocol based on signcryption scheme. The analysis by Kailar logic shows that the protocol can achieve the non-re- pudiation, fairness. Furthermore, it has advantage of resisting the attacks such as distort, replay and conspiracy.

Performance Study of VSAT Slotted ALOHA System Based on Timing Model

Computer Science. 2012, 39 (11): 62-64.

Abstract

PDF(327KB) ( 780 )

RelatedCitation | Metrics

In VSAT satellite communication system using slotted ALOHA method, assigning guard time is important for frame format design. In this paper, sufficient and necessary condition for collision-free receiving in quasi synchronous ALOHA system was reduced from timing relation. Based on this conclusion, the joint effects of guard time and timing warp on system performance were analyzed and a scheme was proposed to calculate optimum guard coefficient. Numeric simulations show that when timing warp coefficient月镇0. 5，there is a optimum guard coefficient app，and using it a guard time achieving maximal throughput can be obtain. When timing warp coefficient月0. 5，increasing frame guard time will decrease system throughput.

Skyline-based Dynamic and Trustworthy Service Selection Method

Computer Science. 2012, 39 (11): 65-68.

Abstract

PDF(423KB) ( 665 )

RelatedCitation | Metrics

Service selection is of crucial importance for the sharing, composition and reuse of services. With rapid growth of the number of candidate services which have same functional attributes but have different non-functional attributes, e. g. quality of service(QoS)，QoS has been regarded as one of the most important criteria for the selection of services. However, the authenticity and trustworthiness of QoS data can not be effectively guaranteed during service selection process in dynamic network environments. To address this problem, trust of service (hoS) including the trust of service resource type and the trust of service provider was introduced. Trust mechanism was therefore established for service selection through the evaluation of two kinds of trust. Association relationship between QoS and ToS was given with the employment of dominance relationship and projection model in Skyline technology. A skylincbased dynamic service se- lection algorithm was proposed to improve the trustworthiness of service selection while QoS requirements arc already met. Simulation results demonstrate the feasibility and effectiveness of the presented algorithm.

Content-based Reversible Steganographic Method for Multimodal Biometrics

Computer Science. 2012, 39 (11): 70-74.

Abstract

PDF(734KB) ( 702 )

RelatedCitation | Metrics

A novel dual reversible steganographic method based on content correlation analysis is proposed for enhan- cing the security of biometric information transmitted in the network. Different from the existing approaches, in order to make good use of abundant content of the cover image and improve the performance of data hiding,least square (LS) re- gression method was adopted to exploit the content correlation between the palm image and the face image firstly. In other words, the face image was used to represent the palm image, and the unrepresented part of the palm image was embedded into the corresponding face image. Besides, the reconstruction coefficients were taken as the secret key for se- curity. hhen, the stego-face image was hidden into a nature cover image by the same method for causing much less at- tendon of attackers randomly. Finally, the stego-cov}r image containing palm information and face information were transmitted. The presented method realizes dual reversible data hiding for biometrics and its security is also employed with a hash function and a secret key. The proposed method exhibits good security, invisibility and high capacity. Spe- cially, dual reversible data hiding further improves the security of transmitted biometric information and ensures the va- lidity for multimodal biometric identification.

Valid Verification of Termination for Some Quantum Programs

Computer Science. 2012, 39 (11): 75-78.

Abstract

PDF(259KB) ( 810 )

RelatedCitation | Metrics

Using the verification method for quantum programs proposed in巨18习，the verification of quantum programs on the single qubit system described by bit flip channel, depolarizing channel, amplitude damping channel and phase damping channel was investigated, and the termination conditions of a duantum program were discussed in detail by se- lecting the different observable operators. It shows that the termination conditions of the duantum program described by quantum channel not only depends on the selection of input state, but also depends on the selection of observable opera- tors.

Baseline Algorithm Based on Adaptive Gaussian Process Machine Learning

Computer Science. 2012, 39 (11): 79-82.

Abstract

PDF(588KB) ( 832 )

RelatedCitation | Metrics

The baseline calculation is an important issue in the field of network monitoring. As to deal with the data, most researches just ignore the probability characteristics of the data,which fails to combine data distribution to predict the data and make the related processing. Therefore, this article analysed the historical data's noise first, then made the prediction with the Gaussian process machine learning, and combined with ant colony algorithm, achieved the adaptive mechanism of the parameters,and then calculated the baselines. The experiment shows that compared with other algo- rithms, our algorithm improves efficiency and accuracy.

Deployment Strategy of Wireless Sensor Network Nodes Based on AFSA-PSO Hybrid Algorithm

Computer Science. 2012, 39 (11): 83-85.

Abstract

PDF(351KB) ( 701 )

RelatedCitation | Metrics

The deployment of sensor nodes was formalized as a combinatorial optimization problem, and the network coverage was used as the objective function. For the model this paper proposed a hybrid algorithm of artificial fish swarm algorithm(AFSA) and particle swarm optimization(PSO) by combining the advantages of the two algorithms. Particle swarm optimization can achieve the effective local search, and artificial fish swarm algorithm can enhance the a- bility of global optimization. The AFSA-PSO hybrid algorithm proposed in this paper has the advantages of both. The simulation results show that AFSA-PSO hybrid algorithm is superior to the artificial fish swarm algorithm and particle swarm optimization algorithm, can effectively improve network coverage with fewer iterations.

Behavioribased .Tunk Mail Detection

Computer Science. 2012, 39 (11): 86-89.

Abstract

PDF(357KB) ( 1184 )

RelatedCitation | Metrics

As an important information exchange tool, emails contribute to serious problems such as spam-mails, mali- cious mails and private information leaks. Junk mail detection is a recently emerging filed,which aims to provide a effi- cicnt monitoring means on deciding whether a mail is a junk mail. We provided a behavior-based method of junk mail de- tection. We showed the behavior models used in our method and gave a introduction of the algorithm processes. We im- plemented our method on a real mail set and gave some comparisons with content based junk mail detection to get some experiences in mining mail information.

Freegate Software Flow Test

Computer Science. 2012, 39 (11): 90-89.

Abstract

PDF(0KB) ( 496 )

RelatedCitation | Metrics

It has happened that someone broke through the Internet filtering and visited the forbidden information on the abroad website in recent years. It has great realistic significance for us to study and control that behavior. Flow classifi- canon technology has always been the research hotspots in the foreign and domestic network measurement direction and we has already obtained very good results in P2P detection area. This document applies the support vector machine tech- nology in the flow classification area to behavior detection of the Frecgate which is a special software used to break the Internet. The results from the tests show that this method has high detection rate on the flow caused by the behavior of breaking Internet and provides a new line of thinking on monitoring the behavior of breaking Internet effectively.

Code Protection Method Oriented to Application Requirement

Computer Science. 2012, 39 (11): 93-97.

Abstract

PDF(407KB) ( 665 )

RelatedCitation | Metrics

Based on compiler analysis and combining handware and software,this paper put forward a protected frame- work in order to prevent attackers to read, analyse and tamper the compiled executable code when the process compiler compiles the source code. Firstly, classifying specific application requirement(data important and algorithm important)， then extracting key code block and generating marked binary code after classification, synthesizing the advantage of digi tal signature(RSA) and (AES) encryption and decryption algorithm, employing digital signature and encryption to en- crypt block marked information,finally loading it to FPGA to verify and run. The experimental analysis show that the method has good maneuverability and maintainability, can reduce the software protection costs and system implement cost, and reach the purpose of protection.

Ontology-based Dynamic Reliability Evaluation Model for Web Services

Computer Science. 2012, 39 (11): 98-101.

Abstract

PDF(404KB) ( 747 )

RelatedCitation | Metrics

Current reliability evaluation approaches lack a formal semantic representation and the support of incomplete or uncertain information. To fill this gap, Web service reliability ontology (WSRO) was developed. WSRO is mapped to a probability graphical model, and on this basis, the Web service reliability evaluation results can be obtained by the eau- sality reasoning. This approach can solve the problem of the lack of semantic description, assess the reliability of Web service, and lay the foundation for Web service auto-discovery, auto-composition and auto-execution. Experimental re- sups show that the approach is feasible,apphcable and effective.

Model Extraction and Reliability Verification on SOCKET Communication Program

Computer Science. 2012, 39 (11): 102-105.

Abstract

PDF(373KB) ( 783 )

RelatedCitation | Metrics

Formal method is a means to verify the reliability and safety of concurrent systems. Formal verification of model which is automatically extracted from concurrent system built from high level language is a hot research topic in the field of model checking technology. With the focus on potential run time problems ( deadlocks, memory leaks, the boundary data loss and other run-time errors) result from abnormal socket function call sequence, this paper analyzed the sequence structure of the socket program, constructed the Promela model of socket functions through the description of message data structures and channels, defined mapping rules of socket function to Promela,proposed the socket func- tion call sequence extraction algorithm and target Promela model generation algorithm, used linear temporal logic (LTL) to describe the property the socket function call sequence shall meet. A socket communication program analysis system was constructed. The experiment result shows that the system can detect the potential run time problems of socket program effectively.

Construction Method of Workflow Model Complex Control Structure

Computer Science. 2012, 39 (11): 106-110.

Abstract

PDF(468KB) ( 679 )

RelatedCitation | Metrics

It is difficult to guarantee correctness of workflow model control structure by artificial, when case semantic of business rules in workflow model changes more complicated. Methods of construction automatically are less. At first well block structured control structure was introduced,which has lots of useful properties in model soundness. I3y ana- lyzing topology of blocks in workflow models, branch and merge control structures and syntax rules they should follow were defined respectively. Algorithms of branch control structure were proposed according to semantics of business rule. Then based on this, algorithms of merge ones were proposed using all branch weights. Finally a practical example was given to verify the constructing process of correct block structure. This method is independent of specific modeling method, so that has three advantages, complete commonality, modeling-independent and wide applicability.

Parallel Implementation Approach for the GeoComputation Service Process in the High Performance Computing Environment

Computer Science. 2012, 39 (11): 111-115.

Abstract

PDF(444KB) ( 1089 )

RelatedCitation | Metrics

GcoComputation is an important advanced computing paradigm in geographic information science. It is an in- dispensable part of the next generation of GIS (Geographic Information System). GeoComputation involves lots of processes such as data processing, spatial analysis, process simulation, etc. hhe process of GeoComputation is so compli- Gated that usually consumes large amount of time. HPC (High Performance Computing) provides a powerful tool to solve the complex GcoComputation problems. To cope with the parallel implementation issues of the GcoComputation service process in HPC envirorunent,a modeling method of the GeoComputation service process and a strategy for pro- cessing concurrent processes based on multithread were proposed. A prototype system of executing engine for the Geo- Computation service process in HPC environment was designed and implemented, which can automatically resolve and dispatch the GcoComputation service process and keep the load balanced in concurrent conditions. hhe experimental re- sups prove that the proposed method can effectively model GeoComputation service process with good practical per- formance,which can be widely used in geospatial data application fields.

Parallel Access Methods for Geographic Raster Data

Computer Science. 2012, 39 (11): 116-121.

Abstract

PDF(526KB) ( 941 )

RelatedCitation | Metrics

In the field of geographic raster data processing, the data I/O performance significantly impacts the perfor- mance of a parallel program, especially when data are massive. Currently, the research on parallel optimizations for geo- graphic raster data I/O is quite limited. I3y combining the data I/O paradigms in parallel programs with the characteris- tics of the geographic raster logical and physical models, we proposed a parallel I/O architecture for geographic raster data, and implemented four parallel access methods based on message passing model. Massive experiments verify that our methods outperform not only the conventional sequential I/O method but also the time-sharing multi-processes data access method. Our work can be used to promote the I/O efficiency of a parallel raster processing algorithm and thus improve the parallel performance of the program.

Improved Possibilistic C-means Clustering Algorithm Based on Particle Swarm Optimization

Computer Science. 2012, 39 (11): 122-126.

Abstract

PDF(414KB) ( 744 )

RelatedCitation | Metrics

An improved possibifistic Gmeans(PCM) algorithm based on particle swarm optimization (PSO) was pre- sented. This algorithm consists of two steps; first, using the improved PCM to calculate the degree of membership ma- trix and cluster centroid to encode particles, which can low the influence of initialized centroid and improve clustering precision. In the second, using PSO to optimize the encoded data points, which can overcome the coincident clusters and avoid easily falling into local optimum The experimental results on the synthetic data sets and UCI data sets show that the proposed algorithm has less computational complexity, higher clustering precision and greater searching capability.

Information-gain-based Text Feature Selection Method

Computer Science. 2012, 39 (11): 127-130.

Abstract

PDF(349KB) ( 1526 )

RelatedCitation | Metrics

Due to the maldistribution of class and feature, the classification performance of traditional information gain algorithm will decrease sharply. Considering that,a text feature selection method TDpIG based on the information gain was proposed. First of all, selected feature in dataset based on the class,which can reduce the effect of dataset imbalance on feature selection. Secondly, calculated information gain weight by using feature occurrence probability to decrease the interference of low frequency words to feature selection. At last, analysed the increasing information of each class by use of dispersion,filtering out the relative redundant features of high frequency words,further refining the selected feature applied increasing information, and getting the uniform and accurate subsets. The comparison experiment shows that the method has better classification performance.

Study on Continuous Bichromatic Reverse k Nearest Neighbor Query Processing in Road Networks

Computer Science. 2012, 39 (11): 131-136.

Abstract

PDF(501KB) ( 996 )

RelatedCitation | Metrics

In recent years,reverse nearest neighbor (RNN) query processing has received wide concern and has become a hot research topic in the database field. Many efficient RNN query processing methods in Euclidean spaces have been proposed. However, there is few related research findings in road networks. RNN processing methods in Euclidean spaces arc unsuitable for RNN processing in road networks where the query-object distance and object object distance arc determined by the connectivity of the road network. RNN query can be roughly classified into two types:Monochro- matic RNN(MRNN) query and 13ichromatic RNN(13RNN) query. Up to now, there is still a lack of efficient algorithms for processing continuous bichromatic reverse k nearest neighbor (CI3RkNN) queries in road networks. Thus, it is es- sential to examine CBRkNN queries in road networks and present efficient methods for CBRkNN query processing in road networks.

Subgraph Similarity Matching Based on Path Mapping

Computer Science. 2012, 39 (11): 137-141.

Abstract

PDF(424KB) ( 1810 )

RelatedCitation | Metrics

So far, the related similarity matching methods usually do not consider the relation between nodes and the practical significance of the weighted edges. This paper presented a similarity subgraph matching method based on path mapping. It can accurately locate the weighted graphs with similar topological structure. `hhe innovations of this paper are to make full use of the attribute information and to consider the topological structure characteristics, overcoming the ignorance of the structure relations and the significance of edge weights to analysis graph similarity. hherefore, the method can greatly improve the application scope of graph similarity matching and the experiment shows that this meth- od has higher searching quality and efficiency.

Automatic Spectral Clustering of User Sessions Based on the Similarity of Global Alignment

Computer Science. 2012, 39 (11): 142-144.

Abstract

PDF(339KB) ( 996 )

RelatedCitation | Metrics

Focusing on the problem of low accuracy of similarity measurement and necessarily determining the number of clustering in advance in clustering algorithms of user sessions in existing personalized recommendation services sys- tans, a global alignment method based on dynamic programming algorithm was proposed to measure similarity between user sessions by integrating the information of serialized visiting pages and visiting times. On this basis,automatic spec- tral clustering was done on the user sessions by using improved NJW clustering algorithm. Experimental results show that the algorithm achieves a higher clustering performance than the comparative algorithms by considering the overall characteristics and local information of user sessions. It can also improve the efficiency of Web personalized recommen- lotion services.

Algorithm to Extract Search Suggested Keyword Based on Apriori

Computer Science. 2012, 39 (11): 145-148.

Abstract

PDF(341KB) ( 771 )

RelatedCitation | Metrics

With the rapid development of Internet, users make use of network to achieve the resources frequently. But search engines return a big and linear list which is uncomfortable for users to use. Next clustering which is flexible and automatic has become one important measure to solve the problem This paper used a kind of clustering algorithm named Lingo as major object And aiming at a problem in Lingo algorithm that the label can't belong to more then one sentence, the paper employed Apriori algorithm to find label as search suggested keyword to make up the problem

Research on Rough K-modes Algorithm with Adaptive Parameters Based on MapReduce

Computer Science. 2012, 39 (11): 149-152.

Abstract

PDF(423KB) ( 684 )

RelatedCitation | Metrics

In the traditional rough K-modes algorithm, three important parameters, w} , wk and。，are set fixedly,making the clustering result unstable and interfered easily by noise. We proposed a rough K-modes algorithm based on MapRe- duce adaptive parameters,which adjusts parameters depending on the characteristics of different clustering stages,opti- mining the clustering result In addition, we designed the rough K-modes algorithm with adaptive parameters that can be used in MapReduce to improve the efficiency of the algorithm and performance of clustering largcscale data. Finally, this algorithm's validity is proved by experiments.

New Algorithm for Mining Workflow Frequent Closet Pattern

Computer Science. 2012, 39 (11): 153-156.

Abstract

PDF(292KB) ( 806 )

RelatedCitation | Metrics

To improve the accuracy of workflow frequent patterns, we proposed a new algorithm for mining frequent closed pattern. First, we extended the definition of the dependence Matrix, which includes direct dependency and over- lapping relationship among activities from the work flow logs. Second, we extended the CHARM algorithm to mine all closed frequent patterns over the dependence matrix Finally, the workflow frequent patterns were generated by discar- ding non-frequent patterns. The algorithm has advantages in dealing with the interleaving relations among activities and workflow models with both serial or parallel relations.

Approximate Duplicate Record Detection Algorithm Based on PSO and LSSVM

Computer Science. 2012, 39 (11): 157-159.

Abstract

PDF(321KB) ( 662 )

RelatedCitation | Metrics

Approximately duplicate record detection algorithm was proposed based on quantum swarm algorithm (QP- SO) and least squares support vector machine (LSSVM) to solve the large-scale database approximation duplicate re- cord detection problem. Firstly, the record field similarity values are calculated, and then the LSSVM parameters are op- timized,by QPSO to construction the approximately duplicate records detection model, finally simulation experiments arc carried out on the data set. hhe simulation results show that QPSC}I_SSVM not only improves the accuracy of the duplicate record detection but also improves the detection efficiency,and it is an effective approximate duplicate record, Detection algorithm.

Approach of Fault Diagnosis Based on Patterns

Computer Science. 2012, 39 (11): 160-164.

Abstract

PDF(434KB) ( 883 )

RelatedCitation | Metrics

An improved method of fault diagnosis based on patterns was proposed. Faults are defined as patterns that are described by regular expressions to avoid the limitation of traditionally defining faults which uses `special' events. Regular expressions can be equivalently transformed into deterministic finite automaton, so that fault pattern space can be easily constructed. All the information of faults can be contained in the fault pattern space, which include fault se- quences, multiple faults, and intermittent faults. Finally, the definition of diagnosability based on fault pattern space was given and a diagnostic system was implemented.

Hybrid Multigranulation Rough Sets Based on Equivalence Relations

Computer Science. 2012, 39 (11): 165-169.

Abstract

PDF(338KB) ( 649 )

RelatedCitation | Metrics

Bascd on the partitions generated from the equivalence relations, the concept of hybrid multigranulation space was proposed to study the multigranulation space with both disjunction and conjunction. 13y using the partitions in hy- brid multigranulation space, the target concept was approximated and then the concept of hybrid multigranulation rough about hybrid set was proposed. Not only the basic properties multigranulation rough set were explored, but also it was proven that hybrid multigranulation rough set is a generalization of both optimistic and pessimistic multigranulation sets.

Structure Identification Method of T-S Fuzzy Model Method Based on Moving Rate

Computer Science. 2012, 39 (11): 170-173.

Abstract

PDF(370KB) ( 714 )

RelatedCitation | Metrics

To improve the effectiveness of the existing fuzzy identification method, a structure identification method based on moving rating was proposed for T-S fuzzy model. The main work is as below. Firstly, the moving rates for S- type, Z-type and trapezoidal membership functions of T-S fuzzy model were defined, and compared with proposed mov- ing rate and the existing grade of the membership function,the proposed moving rate is more effective. Next,T-S fuzzy reasoning method based on moving rating was proposed, and the identification methods for premise and consequence based on moving rate were proposed for T-S model. Finally, the proposed identification method was applied to the fuzzy modeling for the precipitation forecast and security situation prediction. Test results, compared with existing method and fuzzy neural network algorithm, show that the proposed method significantly improves the effectiveness of fuzzy identi- fication, and reduces the number of rule and identification error.

Hybrid Optimized Algorithm Based on Improved MOPSO and Local Search and its Application

Computer Science. 2012, 39 (11): 174-178.

Abstract

PDF(496KB) ( 1267 )

RelatedCitation | Metrics

In order to improve the weaknesses of the particle swarm's easily premature and slow convergence in late stage, the H-MOPSO, based on the integration of improved MOPSO and local search, was proposed. First of all, the non-uniform mutation operator and self-adaptive inertia weight were adopted to enhance its ability of global search. I}hen, the model of MOPSO hybrid with local search was established. According to the model, the local search algorithm based on hill climbing strategy with sidesteps was periodically used to optimize the swarm, making particles search a- long descent direction when they were away from Pareto front, and search along non-dominated direction while they were near Pareto front Simulation results of benchmark functions show that H-MOPSO has better performance com- pared with MOPSO,NSGA-II and MOEA/D. The solving of supplier selection problem further validates its effective- ness.

Proving NP-completeness of Polynomial Reduction from the SAT Problem to the MSP Problem

Computer Science. 2012, 39 (11): 179-182.

Abstract

PDF(304KB) ( 961 )

RelatedCitation | Metrics

According to the MSP problem (defined in the body) raised in paper[1]，this paper started from the SAT problem to the MSP problem. Thus we provided another proof to the NP-completeness of the MSP problem.

Induction of Decision Tree for Continuous-valued Attributes Based on Tolerance Rou沙Sets Technique

Computer Science. 2012, 39 (11): 183-186.

Abstract

PDF(309KB) ( 786 )

RelatedCitation | Metrics

Decision tree is a popular data mining method, and it is a crucial problem to select expanded attributes in the induction of decision tree. I}he uncertainty of each cut of each continuous valued attributes needs to be measured during the selection of expanded attributes for induction of decision tree based on discretion method, and the computational time complexity is very high. In order to deal with this problem, a method of induction of decision tree for continuous- valued attributes based on tolerance rough sets technique was proposed. "hhe method consists of three stages. First ex- panded attributes are selected with tolerance rough sets technique, and then the optimal cut of the expanded attribute is found, and the sample set is partitioned by the optimal cut, finally the decision tree can be generated recursively. We ana- lysed the computational time complexity of the algorithm in theory and conducted some experiments on multiple data- base. The experimental results and the statistical analysis of the results demonstrate that the proposed method outper- forms other related methods in terms of computational complexity and classification accuracy.

Site Deep-digger System Based on Internet Client Psychology Mining

Computer Science. 2012, 39 (11): 187-190.

Abstract

PDF(384KB) ( 738 )

RelatedCitation | Metrics

Site optimization systems based on artificial retrieval and topping work slowly and inefficiently. In order to deal with these problems, a site deep-digger system was proposed with Internet client psychology mining. I}he system u- tilizes Web opinion mining technologies to search Internet hot points. And it can extract and estimate client requests from historic data. According to these rectuests and historic information,it optimizes the site by creating related Webs. Simulation results show that the system can perform better to mine Internet opinions and do SEO (search engine optimi- nation) better than the traditional methods do so.

Length of the Shortest Synchronizing Words for Quasi-trapped Synchronizing Automata

Computer Science. 2012, 39 (11): 191-193.

Abstract

PDF(193KB) ( 831 )

RelatedCitation | Metrics

A synchronizing automaton is said to be quasi-trapped if it is neither trapped nor strongly connected. Let A be a quasi trapped synchronizing automaton. By using the number of the states of the strongly connected sulrautomaton of A,a upper bound of the length of the shortest synchronizing word of A was given,and then a sufficient condition for A to satisfy the Cerny Conjecture was obtained.

incremental Density Weighted Proximal Support Vector Machine

Computer Science. 2012, 39 (11): 194-196.

Abstract

PDF(322KB) ( 684 )

RelatedCitation | Metrics

The proximal support vector machines (PSVM) is a regularized least squares problem, which has an analytic solution. However, the PSVM lacks sparseness, and all training dates become support vectors. I}his paper focused on ef- fectively controling the sparseness of the PSVM. An incremental density weighted proximal support vector machine (IDWPSVM) was proposed, which selects the basis support vectors in the training set, The experiment results show that the accuracy of the IDWPSVM can is similar with the SVM, PSVM and DWPSVM methods, and convergence speed is faster. The )DWPSVM can effectively control the sparseness of the PSVM.

Concept Acquisition of Complement Formal Context Based on the Original Context

Computer Science. 2012, 39 (11): 197-200.

Abstract

PDF(283KB) ( 714 )

RelatedCitation | Metrics

As an efficient tool for knowledge acctuisition,formal concept analysis has been applied to many fields. Based on the extents (intents) of a formal context,some concepts of complement context were obtained by the operator‘+’， and it was proved that these concepts includes all the meet irreducible elements (join-irreducible elements) of this com- plement context. Then, the approach to find all the meet irreducible elements (join-irreducible elements) was given. Fi- nally, all the concepts of complement context were obtained.

Which is More Effective for Chinese Lexical Analysis via Character Tagging:Above-context Versus Below-context

Computer Science. 2012, 39 (11): 201-203.

Abstract

PDF(358KB) ( 689 )

RelatedCitation | Metrics

Chinese lexical analysis is a foundational task for Chinese information processing. At the current, the main- stream technology of Chinese lexical analysis is based on statistical methods. These methods treat the analysis process as a sectuence data tagging problem. Context is the necessary resource not only for obtaining linguistic knowledge in sta- tistical linguistics but also for solving the problem in natural language processing. Chinese lexical analysis needs the help of correlative context. However, are above and below the same important? To overcome the lack of giving the result by the subjective experience,we studied the contribution of above and below for character-based tagging Chinese lexical a- nalysis via the large number of experiments about word segmentation, PUS tagging and named entity recognition. Closed evaluations were performed on many kinds of corpus from the international Chinese language processing 13akeoff, and comparative experiments were performed on different feature templates which describe above-context and below-con- text. Experimental results show that the performance by the below-context increases 6 percentage points than by the a- bovccontcxt.

Artificial Immune Clustering Semi-supervised Algorithm Based on Manifold Distance

Computer Science. 2012, 39 (11): 204-207.

Abstract

PDF(284KB) ( 719 )

RelatedCitation | Metrics

Manifold distance was used as the basic measure of the sample similarity between samples. The pair-wise constrains prior information was introduced,then the measure matrix was obtained through affinity propagation. So the clustering problem was transformed as one optimal model. Clonal selection algorithm was employed to solve this model, and the clustering results were given. Experiments on artificial data sets and UCI benchmark data set show that the pro- posed method can give the better accuracy.

Dynamic Environment Economic Dispatch Based on Differential Evolution Algorithm

Computer Science. 2012, 39 (11): 208-211.

Abstract

PDF(405KB) ( 750 )

RelatedCitation | Metrics

Dynamic environment economic dispatch is of non-linear optimization problems. It represents the characteris- tics of multi-objective,high dimensions and constraints. So the traditional methods are no longer fit to solving these op- timization problems. A price penalty factor approach was utilized here to convert the bi-objective problems into single objective ones. In order to handle constraints effectively,heuristic rules were proposed to handle ramp rate constraints, and heuristic strategics based on priority list arc employed to handle active power balance constraints. The heuristic strategics also can increase the variety of the individual and extend the search scope. I}he thermal unit with the lower average full-load cost will have the higher priority to dispatch more generation power in the heuristic strategies based on priority list, so that the even better scheduling solutions can be obtained. At last, the differential evolution algorithm was improved to enhance the search ability and improve the solution quality.

Pairwise Constraint Projections Inosculating Sparsity Preserving

Computer Science. 2012, 39 (11): 212-215.

Abstract

PDF(373KB) ( 770 )

RelatedCitation | Metrics

A kind of algorithm called Pairwise Constraint Projections inosculating Sparsity Preserving (SPPCP) was proposed,which introduces Sparsity Preserving Projections(SPP) with trade-off parameter in the process of pairwise constraint-guided dimensional reduction,preserving pairwise constraint feature and heriting the special character of geo- metrical structure preserving and neighbor preserving from sparsity preserving. Experiments operated on UCI databases and AR face database show that the algorithm inosculates merits of SPP effectively. Compared with the typical semi su- pervised dimensional reduction algorithms based on pairwise constraint, the algorithm can improve the accuracy and sta- bility of classified algorithms based on the shortest Euclidean distance.

Coal-and-Gas Outburst Forecast Using CCPSO and SVM

Computer Science. 2012, 39 (11): 216-220.

Abstract

PDF(496KB) ( 700 )

RelatedCitation | Metrics

In order to forecast effectively coal-and-gas outburst in coal-mine,a new method for coal-and-gas outburst forecast based on CCPSO (complete chaotic particle swarm optimization) and SVM (support vector machine) was presented. With multi-fractal dimension spectrum of gas emission amount dynamic time series in the front of work-face in coal-mine being feature index, the forecasting model was constructed by using SVM. The parameters vector of the proposed model was selected and optimized by CCPSO and the criteria of CERM (classification error rate and TSSM (test sample set minimization). The experimental results show that the proposed method is effective and provides a new approach for forecasting coal-and-gas outburst in coal-mine.

Search on Granular Logic Based on Qualitative Mapping and its Reasoning Algorithm in Petri Net

Computer Science. 2012, 39 (11): 221-225.

Abstract

PDF(463KB) ( 674 )

RelatedCitation | Metrics

Under cognitive mechanism, the concepts and logical formula of attribute granular was given throuth qualita- live mapping based on the characteristics of uncertainty knowledge processing. Qualitative mapping was established based on the transformation of the relationship between the properties of ctuantitative and ctualitative. The initial granu- lar logic system was constructed on this basis. Finally, the granular logical reasoning was described in the Petri net for- mal methods. I}hc results show that it is effective. It makes the thinking operation that is related to cognitive identifica- lion and judgement can get a better expression.

ALPO-SVM for Road Speed Limit Signs Recognition

Computer Science. 2012, 39 (11): 226-229.

Abstract

PDF(671KB) ( 647 )

RelatedCitation | Metrics

This paper proposed a support vector machine (SVM)-based road speed limit signs recognition method. In order to improve the recognition performance of the algorithm, an adaptive learning particle swarm optimisation (ALP- SO) algorithm was employed to optimize the parameters of SV呱The proposed ALPSO may adjust its learning parame- ters during the evolutionary process to coordinate the local search with the global search of the PSO algorithm. Experi- mental results show that the proposed ALPSC}SVM is superior to the traditional SVM in terms of the recognition rate, and it's convergence performance is better than the standard PSC}SVM.

Date Compression with Homomorphism in Neighborhood Information Systems

Computer Science. 2012, 39 (11): 230-232.

Abstract

PDF(230KB) ( 812 )

RelatedCitation | Metrics

The notion of homomorphism is used as a tool to deal with date compression in neighborhood information systems. We reviewed the notions of neighborhood information systems and the induced neighborhood information sys- terns. Then a novel definition of consistent function was proposed, and some important properties of neighborhood infor- mation systems under homomorphism were discussed. Furthermore, we proved that the reduction of the original system and its image system is equivalent to each other. I3y calculating the reduction of the less data image system, a more con- venient measure to handle the date compression of the original system was obtained.

Study on Acoustic Sources Identification Based on Multi-SVDD

Computer Science. 2012, 39 (11): 233-236.

Abstract

PDF(379KB) ( 677 )

RelatedCitation | Metrics

Traditional acoustic sources identification ways have low recognition rate on abrupt acoustic sources. A new method based on multi-support vector data description to the underwater vehicles acoustic sources identification was proposed.

Local Causal Structural Active Learning Method Based on Causal Power

Computer Science. 2012, 39 (11): 237-242.

Abstract

PDF(510KB) ( 924 )

RelatedCitation | Metrics

Causal structure learning is an important causal knowledge discovery method to disclose the nature of causal interactions in the I3ayesian Networks. The causal relations are difficult to be discovered by only using observation data. On the other hand, actually, we are often only interested in local causal structure about a target variable. This paper presented a local causal structure learning method by integrating feature selection into intervention called a local causal structural active learning based on causal power(CSI-LCSL). CSI-LCSL integrated the dividing structure ability of Markov blanket and causal discovery ability of intervention learning. Firstly, under the faithfulness assumption, CSI- LCSL utilized HITON-MI3 algorithm to obtain the Markov blanket of interested variable for generating a local model. I}hen, we selected a intervention variable from the local model by using non-sys entropy to generate interventional data by perfect experiments. Finally,we used an exact method algorithm to obtain a local causal structure of the interested variable by combining observational data and interventional data. A series of comparative experiments on two standard 13aycsian networks show that our method has excellent learning accuracy.

Research of Image Retrieval Method Based on Color Feature

Computer Science. 2012, 39 (11): 243-245.

Abstract

PDF(844KB) ( 930 )

RelatedCitation | Metrics

Color-based image retrieval has been widely used due to good robustness to various changes of images. How- ever, lack of color spatial information leads to retrieval errors. According to retrieval problems caused by global color histogram and block histogram, the paper proposed a new image retrieval method that uses equal-area rectangle ring to extract image features. Firstly, we made use of equal-area rectangle ring partitioning strategy to block the image. Se condly,we extracted sub-block accumulative color histogram as color features. Thirdly, we set weight values according to gradual reducing of rectangle ring from inside to outside in order to highlight the main part of an image, and got the simi- larity measures of two images by calculating weighed accumulative distance which is between the sulrblocks color fca- lures of two images, then retrieved images and outputed the retrieval results. At last, experimental results show that the proposed method has a higher retrieval effect, comparing to global histogram, cumulative histogram and traditional block histogram.

Fingerprint Orientation Estimation Based on Morphological Operation

Computer Science. 2012, 39 (11): 246-248.

Abstract

PDF(614KB) ( 617 )

RelatedCitation | Metrics

Fingerprint recognition is one of the most popular methods for biometric authentication. In the process of rc cognition, the ridge orientation estimation is the most important step, since it serves for segmentation, enhancement, mi- nutiae extraction and matching. We proposed a method for orientation estimation based on morphological operation to regulate the orientation field. The experience shows that this method can correctly and robustly estimate the fingerprint orientation field for low cauality images, especially for those with creases.

Improved Image Encryption Algorithm Based on Double Scrambling of Fibonacci Transforms

Computer Science. 2012, 39 (11): 249-253.

Abstract

PDF(913KB) ( 778 )

RelatedCitation | Metrics

Image scrambling is one of the important means to realize image encryption and the ultimate purpose of scrambling is to change the pixel position or the gray value, change the statistic information of the image, so as to a- chieve the purpose of image encryption. To enhance the effect and performance of images scrambling, from the perspec- five of uniform blocking,a kind of improved double scrambling algorithm of Fibonacci transforms was put forward. At first, plain image was carried out, using uniform blocking algorithm, and each sub-block pixel was fully diffused to other sub-blocks in turn to weaken the image correlation between adjacent pixels effectively. Next, scrambling algorithm shuf- fled the pixels' position of each blocks to weaken the image spatial correlation by using Fibonacci. Finally, srambling al- gorithm shuffled the pixels' value of the whole image to weaken the iamge color correlation by using Fibonacci. hhe ex- perimental results demonstrate that this algorithm can not only reduce the correlation of image adjacent pixels in essence but also has strong ability against shearing and common noise attacks in some degree, and it improves the encryption system's security by utilizing dual-key cipher.

Realistic Simulation of Three-dimensional Trees Swaying in the Wind

Computer Science. 2012, 39 (11): 254-260.

Abstract

PDF(832KB) ( 1074 )

RelatedCitation | Metrics

This paper presented a highly realistic simulation of the tree swaying in the wind method. The method uses I= system to build the finite element model of tree. Perlin noise function was introduced to establish wind model which has natural wind' s 1/ f fluctuation characteristics. Mechanics of materials beam motion model was used to simulate branches bent by the wind. This paper analysed in detail the trees block to wind: branch block and leaf block, that is "fluid-structure interaction" phenomenon between wind and trec,especially in dealing with the process of branch block, using flow around circular cylinder in the model. The results show that this method gets realistic effects on a tree swa- ying in the wind, and the effect of real-time display can be achieved.

Approach of Moving Object Detection Based on Image Blocks and Improved Particle Filter Algorithm

Computer Science. 2012, 39 (11): 261-263.

Abstract

PDF(619KB) ( 604 )

RelatedCitation | Metrics

In order to detect moving objects quickly and accurately,a method of moving object detection based on image blocks and improved particle filter algorithm was proposed. Firstly, image sectuence was split into blocks and color fea- tures were extracted. I}hen, each image block was processed by the improved particle filter to obtain the weights of par- tides. Finally, according to the average weight of particles, we constructed background model and subtract moving ob- jects. Compared with other detection methods based on single pixel,the proposed method reduces the algorithm compu- tation obviously and improves the execution speed of the code. Experimental results show that this method can accurately and intactly detect moving objects even when faced with repetitive background movement,multi-modal background and illumination changes.

Improved Image Mosaic Algorithm Based on Harris Corner

Computer Science. 2012, 39 (11): 264-266.

Abstract

PDF(582KB) ( 710 )

RelatedCitation | Metrics

In order to solve image mosaic optimization problems, a novel improved algorithm is proposed based on the traditional Harris vertex examination algorithm. The proposed algorithm effectively avoids the affecting of selecting kin the corner response function and the threshold T in detection process. During the process of image mosaic,image regis- tration was improved by NCC. And the mean method was used to fuse spliced images and eliminate stitching gaps. In or- der to verify the effectiveness, the proposed algorithm was evaluated in an example of image mosaic. Simulation results show that this algorithm increases the precision and the efficiency of image mosaic. Consectucntly, this algorithm can ef- ficicntly achieve a goal of seamless mosaic and satisfy the actual demand.

Color Image Segmentation SVM Approach Based on Training Samples Automatic Selection

Computer Science. 2012, 39 (11): 267-271.

Abstract

PDF(671KB) ( 909 )

RelatedCitation | Metrics

Image segmentation is an important research field of pattern recognition, image understanding and computer vision. Support vector machine (SVM) is now widely used in image segmentation, but the training samples are usually selected artificially. hhis will reduce the self-adaptability and affect the classification performance of image segmenta- lion. This paper presented a color image segmentation SVM approach based on training samples automatic selection. First,Fuzzy C-Means (FCM) clustering algorithm was used to obtain the training samples for SVM automatically. hhen, color and texture features were extracted from the image as attributes of training samples of SVM. Finally, the images were segmented by the trained classifier. The experiment results demonstrate that the proposed approach can a- chieve good segmentation performance.

Adaptive Segmentation Algorithm of Visual Impurity in Liquid Based on Motive Information

Computer Science. 2012, 39 (11): 272-276.

Abstract

PDF(692KB) ( 646 )

RelatedCitation | Metrics

The segmentation of motive impurity in liquid plays a key role in automatic detecting system. A fuzzy adap- five threshold segmentation algorithm of visual impurity in liquid based on spatial block variance and its gradient was proposed. At first, the two difference images were partitioned into 5 X 5 blocks, and then the product of the homologous block variance and its gradient of the two difference image were chosen as the block feature. After that, an adaptive threshold was used to segment the image. To be adapted to the change of the block grey value,a fuzzy inference method was adopted to adjust the threshold adaptively according to the block variance and it s gradient. hhe experimental and factual testing results show that the proposed algorithm can meet the demands of visual impurity in liquid in real-time detecting, and that it is a practicable and effective image segmentation method.

Miscanthus Sacchariflorus Cells Image Segmentation Based on Curvelet Transformation

Computer Science. 2012, 39 (11): 277-279.

Abstract

PDF(252KB) ( 664 )

RelatedCitation | Metrics

In order to extract the cellulose of Miscanthus sacchariflorus cells, the fast discrete curvelet transform combi- ning with gray level co-occurrence matrix was applied to extract the image feature vectors, and then SVM was used to segment the images. Experiments arc carried out with this method, and the results show that it is effective to compute the gray level co-occeurrence matrix of sub-band coefficient with”Angle second order moment"，”contrast"，”correla- tion" and”entropy" as the four statistics in image transform domain. According to this, texture segmentation on Miscanthus sacchariflorus cells is feasible. And compared with the texture segmentation method based on gray level co- occurrence matrix(ULCM) , the new method shortens the running time and makes the segmentation accuracy improved.

Based on the Distance of the Rigid Body Murkowski for 3D Movement Recovery Algorithm Research

Computer Science. 2012, 39 (11): 280-282.

Abstract

PDF(230KB) ( 655 )

RelatedCitation | Metrics

The existing non rigid body three dimensional motion reconstruction algorithm is difficult to find different shapes according to different scene and different non rigid body mus, and this kind of situation will exert great influence on the reconstruction process, cause model distortion. In order to solve this problem, this paper put forward a murkowski distance based continuous non rigid body three dimensional movement recovery algorithm, which presents continuous physical characteristics on the decomposition of image sequence of frame and frame between the movement parameters and feature point displacement change. Using the physical characteristics will solve motion structure parameter in Murkowski range bound through the nonlinear optimization method, and get the non rigid body 3D motion structure, and the experiment shows that simulation verification results arc feasible.

Fast Flame Detection with GMM

Computer Science. 2012, 39 (11): 283-285.

Abstract

PDF(834KB) ( 663 )

RelatedCitation | Metrics

Flame Detection based on videos is important for research on fire prevention. I}o improve the detection per- formance,GMM(Uaussian Mixture Model) which can adapt the change of background was used to detect moving sub- jects at first. Then the features of color and changing area were extracted from these moving objects. Finely, these fea- tures were applied to recognize whether there is a flame or not in the scene. The method can not only effectively detect flame in videos, but also save computing time in non-fire scene.

Image Processing Algorithms Based on Linear Algebra

Computer Science. 2012, 39 (11): 286-288.

Abstract

PDF(220KB) ( 688 )

RelatedCitation | Metrics

Upon the theory of 13aycsian algorithm and image processing technique, this paper put forward Markov Chain Monte Carlo Methods, to process the images. The simulation results show that the I3ayesian-Markov Chain Monte Carlo Method fusion algorithm needs only Gibbs sampling 10% iterations and relatively simple j ump nuclear. I}his method has high efficiency and low implementation complexity.

Leaves Recognition System Based on Texture Modeling

Computer Science. 2012, 39 (11): 289-290.

Abstract

PDF(144KB) ( 643 )

RelatedCitation | Metrics

A leaves recognition system based on texture modeling was set up. Firstly, we extracted the important region from the collected leaves,used wavelet transform to obtain the high frequency component,and then used general Gaus- sign distribution to model the high frequency component, used the parameters of the general Gaussian distribution as the vector of the tree. I}he classifier based on KNN was used to classify the pattern. I}he simulation demonstrates the effi- ciency and accuracy of this system.

Compensation Method for Texture Mapping Using Casual Images

Computer Science. 2012, 39 (11): 291-293.

Abstract

PDF(506KB) ( 670 )

RelatedCitation | Metrics

Thc local photogrammetry texture mapping using casual images is a new 3D model texture mapping method, but there is texture distortion from lack of local texture. A compensation method of texture mapping was presented. On the basis of the local photogrammetry and constraint parameters texture mapping method, according to the relationship between the model and casual images, the area of texture missing is determined. I}he minimum rigid-body dynamics iner- tia determination method and mirror image method are utilized to search for the symmetry plane within the scope of model mapping. I}he texture information of the symmetric points is got to implement compensation of texture mapping for the area of texture missing. Experimental results show that,for the local symmetrical 3D modcl,the area of texture missing is complemented,and the texture stretching distortion is eliminated.

Modeling-based Performance Evaluation Methodology of Network on Chip

Computer Science. 2012, 39 (11): 294-297.

Abstract

PDF(351KB) ( 1294 )

RelatedCitation | Metrics

The entire design space is extremely large,and there is huge difference of performance between different op- dons. So performance evaluation of different design becomes an important issue in initial stages. The paper took mode ling-based performance evaluation methodology as masterstroke, discussed research status and metrics of evaluation methods which include simulation modeling and analysis modeling, and summarized accuracy estimating methods of NoC evaluation models. Also, a comparison was made between methods mentioned above. Finally, a guidance of future re- search was proposed based on the analysis of future challenges.

New Test Ggeneration Scheme Based nn Serial Shifter

Computer Science. 2012, 39 (11): 298-300.

Abstract

PDF(328KB) ( 694 )

RelatedCitation | Metrics

As we known, the test of integrated circuits(IC) always encounters a lot of problems, like, processing the huge test data volume, spending a long time to test, and constructing the complex test structure. The paper proposed a serial shifter method for test generation based on extension scan chains. It is on basis of F八N algorithm and makes use of don't care bits `X' of test set. Scan chains can generate test vector by itself. Experiments were made for the whole scan chain and the segmentation. Results show that the proposed method can achieve less test data volume compared with some other classical method. The whole shift and the segmentation one arc prevailing in compression rate and test time respectively.

Modeling and Verifying Scheduling Algorithm TDS for Multiprocessor Systems

Computer Science. 2012, 39 (11): 301-305.

Abstract

PDF(337KB) ( 1033 )

RelatedCitation | Metrics

In multiprocessor systems, the tasks which need to be completed by an application can be distributed onto the same processor or a group of processors. Therefore, software testing is unable to verify the reliability of multiprocessor scheduling algorithms. Thus a formal model based on extended I3iichi automata which is used to describe the丁DS CTask Duplication based Scheduling) algorithm was presented. I}he expected properties of the algorithm were described by 1i near temporal logic formulas and verified on the above model. This method overcomes the limitations of software testing and can be used to check the reliability of the multiprocessor task scheduling I’DS.