Table of Content

01 July 2009, Volume 29 Issue 07

Previous Issue    Next Issue

Pattern recognition and Software

New feature selection approach(CDF) in text categorization

2009, 29(07):  1755-1757. 

Asbtract ( )   PDF (498KB) ( )  

Related Articles | Metrics

Reducing the high dimension of feature vectors is an essential part of text categorization. After studying current dimension reduction technique and analyzing some normal methods of feature selection, a new approach, named CDF, in feature selection was proposed by comprehensively taking concentration among classes, distribution in class and average frequency in class into account. Experiment takes K-nearest neighbor(KNN) as the evaluating classifier. Experiment results prove that CDF approach is simple and effective, and get a better performance than conventional feature selection methods in dimension reduction.

Reduction in ordered decision system based on spearman rank correlations

2009, 29(07):  1758-1759. 

Asbtract ( )   PDF (475KB) ( )  

Related Articles | Metrics

Reduction is an important knowledge discovery process. The original rough set theory based on equivalence relations does not consider attributes with preferenceordered domains and is sensitive to data noise. Based on spearman rank correlation analysis, a method was introduced to reduction ordered decision systems. This method computes the reduction according to the impact of certain attributes on the spearman rank correlations between objects. Experimental results show that this method is effective as it not only takes into account the preference relationship, but also improves noise immunity, and thus is more suitable for practical application.

A workflow-based method for customizing J2EE legacy systems

2009, 29(07):  1760-1763. 

Asbtract ( )   PDF (657KB) ( )  

Related Articles | Metrics

Focused on the popular J2EE platform, this paper gave out a workflow based method named JSPMP, to easily customize the business processes in legacy J2EE systems. JSPMP makes use of the process definition and task management abilities of workflow engines, so that it can control and customize business processes in a dynamic way. Proved by practices, JSPMP is an agile method to customize business processes in popular J2EE legacy systems, and the cost of customizing is also largely reduced.

Development of Ajax component based on JSF custom lifecycle

2009, 29(07):  1764-1766. 

Asbtract ( )   PDF (623KB) ( )  

Related Articles | Metrics

In order to improve the clients’ interactivity of the JSF component, a framework was presented to develop Ajax component on the basis of analyzing the implement mechanism of the custom lifecycle in JSF. By designing the Ajax event class and introducing the related custom lifecycle object in this framework, the processing of the Ajax event could be encapsulated inside the JSF component, just like the normal Faces event. The practical applications indicate that this framework can take the respective advantage of JSF and Ajax and Web user interface can be developed to be more rapid and efficient.

Design and implementation of program mining system framework for active service

2009, 29(07):  1767-1770. 

Asbtract ( )   PDF (647KB) ( )  

Related Articles | Metrics

To solve the problem of Web service about intelligence and personality, a program mining system framework was designed and implemented through the introduction of domain ontology and user interest. Component was described using ontology and feature, and a component retrieval framework based on multirepository and an algorithm for retrieving associated components were presented. Then a component retrieval system was given, and an instance with components in E-Commerce field proved validity compared with the retrieval methods based on keywords and facet. The experiments show that the method has higher precision and recall than others for large-scale components library.

Query scheme and its optimization for distributed data stream

2009, 29(07):  1771-1774. 

Asbtract ( )   PDF (613KB) ( )  

Related Articles | Metrics

Stream database systems are a new type of database system designed to facilitate the execution of queries against continuous streams of data. Many stream-based applications are inherently distributed, Since input stream rates and other system parmeters such as the amount of abailable computing resources can fluctuate significantly, a stream query plan must be able to adapt to these changes. Propose a distributed stream query plan and optimized, use tuple response time and system throughput to evaluate the performance of plan. Compare with other plan by experiments.Proved this plan is optimal.

Security and efficiency improvement for image encryption algorithm based on high-dimension chaotic system

2009, 29(07):  1775-1778. 

Asbtract ( )   PDF (880KB) ( )  

Related Articles | Metrics

Concerning the highdimension chaotic image encryption algorithm proposed by Sun et al., the authors analyzed the potential flaws and proposed corresponding improving methods in terms of security and efficiency respectively. Theoretical analysis and simulation experimental results testify that the enhancement scheme can maintain all the advantages of the original one and conquer the potential flaws. Moreover, the final scheme only needs two encryption rounds to achieve satisfactory performance.

Efficient identification scheme with bilinear map

2009, 29(07):  1779-1781. 

Asbtract ( )   PDF (499KB) ( )  

Related Articles | Metrics

Concerning the discrete logarithm problem of elliptic curve and bilinear mapping, an identification scheme was proposed, and then it was extended to multiKGC situation. The proposed scheme required less computation and communication than the existing ones and realized mutual authentication. Given the condition of the intractability of the elliptic curve discrete logarithm problem, the proposed scheme was proved secure against forged user attack, impersonation attack and replay attack.

Anti-cheating visual cryptography scheme based on probability

2009, 29(07):  1782-1784. 

Asbtract ( )   PDF (455KB) ( )  

Related Articles | Metrics

An anti-cheating Visual Cryptography Scheme (VCS) scheme based on probability, which can find the cheaters without extra information, was proposed. Meanwhile, the visual effects of recovered secret image could be controlled in this scheme. Performance analysis and simulation results show that this scheme has little pixel expansion and the visual effect of recovered secret image can be adjusted by controlling probability.

Anonymous divisible E-cash protocol based on PBOC IC card

2009, 29(07):  1785-1789. 

Asbtract ( )   PDF (817KB) ( )  

Related Articles | Metrics

In this paper, a new anonymous divisible E-cash protocol was proposed, by making use of the physical security of PBOC2.0 IC Card, to support offline transaction. The proposed protocol is more efficient than those based on partial blind signature or zeroknowledge proof. It also integrates a control mechanism to regulate withdrawal, payment, loss reporting and deposit of coins, so as to erase the relationships among different coins and maintain anonymity. Moreover, security was proved under both cryptology principles and financial security standards. This affords the protocol to be more suitable for Internet transaction, offline transaction and P2P transaction. Especially, it can act as a practical PBOC IC card application complement.

Effective strategy to prevent click fraud

2009, 29(07):  1790-1792. 

Asbtract ( )   PDF (486KB) ( )  

Related Articles | Metrics

The search engine companies and pay-per-click model have been questioned because of click fraud. Click fraud has become one of the major diseases in the online advertising industry. Aiming at the difficulties faced by the online advertising industry, a graphics-based authentication code to prevent click fraud strategy was proposed. The strategy can shield effectively software click fraud similar to the clicker, and shield effectively viewers unconscious accidental invalid clicks, and reduce significantly the efficiency of artificial click fraud.

Research on the secure scheme of manufacture design documents based on PKI

2009, 29(07):  1793-1795. 

Asbtract ( )   PDF (485KB) ( )  

Related Articles | Metrics

To cater for the special security requirement for the transmission of product design documents in network-based manufacturing systems, a secure scheme is proposed based on PKI. The scheme adopts symmetric encryption technology to assure the efficiency of the encryption and decryption of the design drawing, and realize the safe transmission of the symmetric key among multiple users by data-envelop technology; the sanitizable digital signature technology is used to documents in VRML format and the authentication of partial file is realized without the whole file being demonstrated, which solves the contradiction between invisibility and integrity. System based on B/S mode and the general VRML format has a good adaptability and cross-platform.

Covert channel detection in ICMP payload based on information entropy SVM

2009, 29(07):  1796-1798. 

Asbtract ( )   PDF (458KB) ( )  

Related Articles | Metrics

ICMP traffic is often considered to be benign，so network equipment does not check its payload．Therefore，an attacker can hide any generated information in the payload of ICMP. The paper analyzed the relation between larger entropy value of the data plane and the support vector machine data，and a new information entropy support vector machine model was proposed．Preliminary experimental results show that the information entropy support vector machine performs well in classification speed and accuracy．

DDHP-based multi-signature scheme with tight security reduction

Zecheng Wang

2009, 29(07):  1799-1802. 

Asbtract ( )   PDF (672KB) ( )  

Related Articles | Metrics

An efficient multisignature scheme was proposed, whose security is tightly related to the hardness of the Dicisional DiffieHellman problem (DDHP) in the random oracle model. Under the Proof-Of-Possession (POP) key registration assumption, the scheme was proved unforgeable against the adaptive chosen message and signing groups attacks. Both the size of multi-signature and the computation cost of verification do not increase with the increase of the signers. Compared with the Computational Diffie-Hellman Problem (CDHP)-based scheme with tight security reduction proposed by Z C WANG, the scheme has the merits of fewer communication rounds and less computation cost. Thus it can be used in many electronic transactions.

Enhanced Rabin cryptosystem based on cubic congruence equation

2009, 29(07):  1803-1805. 

Asbtract ( )   PDF (411KB) ( )  

Related Articles | Metrics

The Rabin cryptosystem was improved and enhanced by substituting quadratic congruence equation with cubic congruence equation in the enciphering and deciphering process.Then, higher security was achieved without increasing the computational complexity. For a special but significant case, all the solutions admitting analytical formulations were found. The obtained results were supported and verified by some numerical examples. Motivated by this theory, a new oblivious transfer protocol was advanced, which offered an optional transmission success rate.

Application of unbalanced data approach in high-speed network intrusion detection

2009, 29(07):  1806-1808. 

Asbtract ( )   PDF (611KB) ( )  

Related Articles | Metrics

In view of the current problems of high-speed network intrusion detection system, such as high packet loss rate, slow pace of testing for attacks and unbalanced data for detection, this paper proposed a new two-stage strategy with load balancing intrusion detection model. In the on-line phase, the system captured the packets from network and split into small ones according to the protocol type, and then detected through each sensor. In the off-line phase, training dataset was used to build module which can detect intrusion. The authors discussed different approaches to unbalanced data, empirically evaluated the SMOTE over-sampling approaches and classified with AdaBoost and random forests algorithm. The experimental results show that SMOTE and the AdaBoost Algorithm by using random forests as weak learner not only can provide better performance to small class,but also has steady model building time.

Research on permission revocation in dynamic security policies

2009, 29(07):  1809-1812. 

Asbtract ( )   PDF (649KB) ( )  

Related Articles | Metrics

In this paper a revocation mechanism based on VTC and in-progress cross list was presented. The revocation is implemented by checking the validation and the timestamp of permissions and a two-way searching on the in-progress cross list on L4 microkernel. Experimental results show that the mechanism is effective and only increases approximately 1% system overhead.

Implementation and performance analysis of signcryption scheme in wireless sensor networks

2009, 29(07):  1813-1815. 

Asbtract ( )   PDF (613KB) ( )  

Related Articles | Metrics

Signcryption is a cryptographic method that fulfills both the functions of secure encryption and digital signature with less cost. To transmit secure and authenticated message in wireless sensor networks, an efficient signcryption scheme with forward secrecy based on elliptic curve, which was presented by HWANG, was modified and it was implemented in TinyOS, which is called TinySignCrypt. Compared with other schemes, the experimental results in the MICA2 platform show TinySignCrypt has better performance in terms of communication and computation load.

A random pair-wise key pre-distribution scheme based on LU matrix space

2009, 29(07):  1816-1819. 

Asbtract ( )   PDF (616KB) ( )  

Related Articles | Metrics

In this paper, we give the detail analysis of a pair-wise key pre-distribution scheme based on LU-decomposition, presenting several security issues including uneven distribution of key information, disclosure of U matrix and the fact that size of system has the great impact on efficiency; Then we propose an improved scheme based on matrix space which combines random scheme and LU-decomposition scheme, the proof of its feasibility and security are also given; Besides, according to the results of time test on PC and SunSpot devices, we conclude that our scheme reduces storage and computation compared to the original scheme.

Schedulability verification of AADL model based on UPPAAL

2009, 29(07):  1820-1824. 

Asbtract ( )   PDF (809KB) ( )  

Related Articles | Metrics

A formal analytical and verification method was proposed to solve the schedulability problem of Architecture Analysis and Design Language (AADL) model. It used model checker UPPAAL to model external environment and to verify AADL thread components schedulability under nonpreempt dispatch strategy. Moreover, a tool for models translation from AADL to UPPAAL was implemented. Experiment demonstrates that analyzing and verifying the schedulability of AADL models by UPPAAL is feasible. This method produces more precise results than other informal verification method.

Private key updating scheme of identity-based public key cryptography under distributed network

2009, 29(07):  1825-1827. 

Asbtract ( )   PDF (438KB) ( )  

Related Articles | Metrics

A new private key updating scheme of IDPKC under distributed network was proposed. The new scheme allotted all updating requests to the entire updating period by introducing a mapping function, and avoided the congestion of updating requests. The paper also introduced the private key updating process and the session key agreement process under the new scheme. Analysis shows the new scheme can fulfill the security requirement and has high expansibility.

Safety requirement oriented test sequence generation algorithm for safety communication protocol

2009, 29(07):  1828-1831. 

Asbtract ( )   PDF (766KB) ( )  

Related Articles | Metrics

The testing of safety communication protocol is the most important part in the simulation testing of safety-critical system, in which generating an effective and comprehensive sequence acts as the key issue. In this paper, a new algorithm to generate safety test sequence was proposed, and definitions of safety factor and safety model of EFSM were given. Compared with traditional algorithms, the new method performs better and achieves higher quality of safety testing. The proposed method has been successfully applied in safety simulation testing of communication protocol in train control system.

Image encryption based on chaotic system with variable parameters

2009, 29(07):  1832-1835. 

Asbtract ( )   PDF (861KB) ( )  

Related Articles | Metrics

Concerning the degeneracy of the computer limited precision effect in a single onedimensional discrete chaotic system, this article proposed a method of pseudorandom sequence generation with variable parameters in the process of the generation of the chaotic signal. The constructed chaos system based on this method had some advantages, such as the long cycle of pseudo-random sequence, the large quantity of key and the big superiority of key space, and the prodeced password has a higher safety performance. Besides, based on the method, a new image encryption algorithm was put forward. The simulation results show that this algorithm is simple, safe and easy to meet the software and hardware requirements.

A safe mental poker protocol with two parts under hostile model

2009, 29(07):  1836-1838. 

Asbtract ( )   PDF (548KB) ( )  

Related Articles | Metrics

Secure Multi-Party Computation is an important research field in cryptography, playing an extremely important role in guaranteeing the information security. The mental poker game utilizes the network to play poker, which can be regarded as a kind of special Secure Multi-Party Computation. Combining the advantage of Hyper-elliptic Curve Cryptosystem with the relevant demands for the mental poker game, this article puts forward a safe mental poker protocol with two parts under hostile model, proves the exactness of the protocol, analyses the security of the protocol under hostile model, and provides calculation complexity and communication complexity of the protocol.

Fairness analysis of electronic payment protocol based on offline-TTP

2009, 29(07):  1839-1843. 

Asbtract ( )   PDF (819KB) ( )  

Related Articles | Metrics

The electronic payment protocol based on offline-TTP is made up of multiple sub-protocols, and is significantly different from traditional authentication protocol, especially on protocol goal and penetrator model. The main purpose of the electronic payment protocol is to realize fair exchange between both buyers and sellers who are assumed to be dishonest. That is, it is necessary for those buyers and sellers to prevent the internal attack. Based on redefining protocol fairness and penetrator model, an extended strand space model which takes an electronic payment protocol as an object is proposed to demonstrate a fairness formal analysis method. The security vulnerabilities and improvement suggestions of the electronic payment protocol are also discussed.

Dynamic buffer partitioning and replacement for reliable multicast

2009, 29(07):  1844-1848. 

Asbtract ( )   PDF (886KB) ( )  

Related Articles | Metrics

ACP is a buffer management protocol applied to shared environment. However, when multicast group number or group member changes, ACP can hardly adjust its allocation quota in time and this will lead to the decrease of cache utilization ratio. To address these drawbacks, a new buffer management scheme called Adaptive and Active Reliable Multicast Protocol (AARM) was proposed, for largescale multicast network. Simulation results show, AARM outperforms ACP in reducing recovery latency and bandwidth consumption, and raising network throughput.

Algorithms for channel assignment in multi-channel wireless mesh networks

2009, 29(07):  1849-1851. 

Asbtract ( )   PDF (449KB) ( )  

Related Articles | Metrics

Aiming at the bandwidth capacity problem in the wireless mesh networks, based on wireless network interfere protocol model for measuring wireless link interference, the paper formulates the channel assignment problem of finding the channel assignment scheme as integer linear program and proposes a channel assignment heuristic algorithm by using objective function for optimizing channel assignment in order to minimize the overall network interference weight during assigning channels to wireless communication links in the wireless mesh networks. Simulation results show that network throughput can be improved significantly by the proposed algorithm.

Mobile Agent-based distributed interactive method in ubiquitous computing environment

2009, 29(07):  1855-1857. 

Asbtract ( )   PDF (516KB) ( )  

Related Articles | Metrics

In order to enhance the transparency and persistency of humancomputer interaction in ubiquitous computing environment, this paper presented an Agentbased distributed interactive method, by which the users communicated with each other through interactive Agent realized by mobile Agent. The interactive Agent acting as interactive entities on behalf of the users migrated and maintained the states of interaction as required in the network, and implemented different interactive applications based on the various kinds of customizable behaviors and collaborative operation modes. The proposed method can achieve transparent, smart, mobile interaction process among the interactive entities, and provide timely interaction services and flexible information sharing manner.

Double-granularity connection admission control scheme based on compression sensing

2009, 29(07):  1858-1860. 

Asbtract ( )   PDF (480KB) ( )  

Related Articles | Metrics

Data compressing techniques are often adopted in practical network systems to improve the efficiency of traffic transmission. However, compression status is ignored in most admission control schemes that usually measure the performance in term of blocking (dropping) probability. In this paper, a Double Granularity Multiple Threshold (DGMT) CAC scheme based on compression sensing was proposed. By sensing compressing levels of arrival connections, DGMT scheme controlled connection admission in granularities of both connection and original information. Multiple thresholds strategy was used in DGMT. Alternatively, Original Information Loss Ratio (OILR) and Transmission Efficiency of Original Information (TEoOI) were presented to measure CAC performances. Theoretical analysis and simulation results show that OILR and TEoOI can well indicate CAC performances, and the performance obtained by compression sensing CAC algorithm improves significantly.

Research on multi-threading processing of concurrent multiple continuous k-nearest neighbor queries

2009, 29(07):  1861-1864. 

Asbtract ( )   PDF (642KB) ( )  

Related Articles | Metrics

To deal with the multiple concurrent continuous k nearest neighbors queries towards moving objects, the proposed a Multithreading Processing of Multiple Continuous Queries (MPMCQ) framework, which adopted pipeline strategy and departed the continuous query processing into three simultaneous stages: query processing, query executing and query results dispatching to improve the parallelism with multithreading technology. Based on MPMCQ framework and grid index for moving objects, a Multithreading processing of Continuous k Nearest Neighbor queries (MCkNN) algorithm was also presented. Experimental results and analysis show that MCkNN algorithm outperforms CPM, YPK-CNN,etc.

Design and Realization of the home robot control in WiFi Mobile IP network-Based on PDA

2009, 29(07):  1865-1867. 

Asbtract ( )   PDF (550KB) ( )  

Related Articles | Metrics

This paper presents a novel approach that utilizes PDA to control the home robot through WiFi Mobile IP network. A new software, named Robot PDA Teleoperation, for controlling the home robot is developed, which can be run on PDA and intellectual mobile phone platform. This approach can overcome the existing deficiency of the simpleness of control, high cost, inconvenience in operation and so on. A number of experiments were carried out based on Canada DrRobot X80-H robot. The experimental results show that the scheme and our software are feasible and effective.

Improvement strategy of MAC for asymmetric link

2009, 29(07):  1868-1870. 

Asbtract ( )   PDF (601KB) ( )  

Related Articles | Metrics

The existing routing protocols for wireless Ad Hoc networks are based on the hypothesis of symmetric link, which cannot adapt to the prevalent asymmetric link. The authors proposed a project named RMAC to improve the IEEE 802.11 protocol. It solved the above problem by changing the format of frames, and doing branch processing to received frames. Then it solved the problem of asymmetric link by normal transmit mechanism instead of setting tables of direct and indirect neighbors. Emulation results verify that RMAC improves network performance without increasing the system overhead.

Clustering algorithm based on group mobility for tactical Ad Hoc networks

2009, 29(07):  1871-1873. 

Asbtract ( )   PDF (670KB) ( )  

Related Articles | Metrics

How to reasonably keep clustering is the key work of tactical Ad Hoc network. It can help Ad hoc network to realize hierarchical network structure and clustering algorithm in military domain. According to the core work, the character of group mobility was introduced. Considering node's local mobility and energy consumption, this paper proposed a clustering algorithm based on group mobility for Tactical Ad Hoc networks and its clustering performance was verified through simulation analysis.

Run-time prediction algorithm based on chunk in grid

2009, 29(07):  1874-1876. 

Asbtract ( )   PDF (607KB) ( )  

Related Articles | Metrics

Aiming at the issues of no real time and not sensitive to load changing, this paper proposes a new chunk prediction algorithm for parallel tasks in grid. Considering the characters of task execution on grid nodes, this algorithm divides the task into two parts and uses the execution time of the first part to predict the remaining part. The experiments results show that our method can achieve more efficiency and generality.

Study of electrical soliton as transmission pulse for UWB communication

2009, 29(07):  1877-1879. 

Asbtract ( )   PDF (604KB) ( )  

Related Articles | Metrics

This paper introduced soliton phenomena in nonlinear transmission line, analyzed the principle of electrical soliton and electrical soliton oscillator, and also gave a simulation experiment under the restrictions of bandwidth and emission power permitted by Federal Communications Commission （FCC）. The simulation results show that the electrical soliton signal can be used as transmission pulse for UWB.

DNA computation for Chinese Postman Problem

2009, 29(07):  1880-1883. 

Asbtract ( )   PDF (634KB) ( )  

Related Articles | Metrics

Such new concepts as virtual weight and virtual node were proposed, and based on which, a novel algorithm based on DNA computation was designed to solve the Chinese Postman Problem (CPP). In the new proposed algorithm, all the feasible solutions of the CPP were obtained by utilizing the Polymerase Chain Reaction (PCR) techniques to eliminate false solutions from all the possible solutions, and then, the optimal solutions were extracted from those feasible solutions by utilizing surface-based techniques for DNA computation and fluorescencelabeling. The results indicate that the new algorithm has good characteristics such as simple encoding and readability.

Intelligent test paper generation based on improved genetic algorithm

2009, 29(07):  1884-1886. 

Asbtract ( )   PDF (505KB) ( )  

Related Articles | Metrics

The test paper generation is a multiconstraint and multiobjective optimization issue. A new mathematical model of intelligence test paper generation system was set up. To avoid slow-convergence and local convergence of simple genetic algorithm (SGA), a kind of improved genetic algorithm was proposed in this paper. The experimental results show that the new method is more efficient and effective to deal with the problem of intelligent test paper generation.

Partheno-genetic algorithm based on dual fitness

2009, 29(07):  1887-1889. 

Asbtract ( )   PDF (446KB) ( )  

Related Articles | Metrics

In this paper a new partheno-genetic algorithm was proposed. It calculates the fitness of individuals and genes respectively, and changes the worst gene. It has greatly improved the global convergence of genetic algorithm (GA). The simulation results show that the new algorithm not only can search the global optimum solution in a short time，but also has excellent stability．

Fuzzy SVM with entropy membership and ant colony optimization

2009, 29(07):  1890-1893. 

Asbtract ( )   PDF (593KB) ( )  

Related Articles | Metrics

Abstract: Compared with traditional support vector machine (SVM), Fuzzy support vector machine (FSVM) made a great improvement in sensitive to outliers or noises. By selecting a proper clustering center to calculate membership, FSVM could obtain higher precision optimization hyperplane. Based on FSVM, through improving clustering center and membership, we propose a new fuzzy clustering method—Fuzzy SVM with Entropy Membership and Ant Colony Optimization(EAFSVM). Experiments show EAFSVM has a better precision and classification performance, especially to multi-class and large scale dates.

Improved BP algorithm based on dynamically adjusting learning rate

2009, 29(07):  1894-1896. 

Asbtract ( )   PDF (398KB) ( )  

Related Articles | Metrics

The learning rate is usually invariable in basic Back Propagation (BP) algorithm，thus the constringency rate and stabilization of network are constrained. Therefore, a BP algorithm based on dynamically adjusting learning rate was proposed. The average absolute error between actual output value of the output layer node and the expected output value and its change ratio were regarded as independent variables, then the function relation of learning rate and two independent variables was found. According to the actual learning circumstance of network, the learning rate was adjusted dynamically. Through the instance simulation, the improved BP algorithm is of more fast constringency rate while keeping the good stabilization than basic BP algorithm. Further more, the algorithm can select the appropriate number of parameters without any condition, and it is therefore of general applicability.

Application of data confusion algorithm in ZigBee protocal

2009, 29(07):  1897-1900. 

Asbtract ( )   PDF (633KB) ( )  

Related Articles | Metrics

Related specialization of data fusion is not considered in Standard ZigBee protocol and this limits its application to low data redundancy. Data redundancy of mess network is high, which will lead to frequent channel competition, time delay of network and network paralysis. Twolayer data confusion was proposed to solve this problem; the first layer is on statistical confusion between terminal nodes and router nodes, and the second layer is on artificial neural network between router nodes and gateway node. Experiments prove data confusion is helpful to reduce data redundancy, improve query effect and reduce energy loss, and make ZigBee protocol suitable for high data redundancy.

Algorithm for visualization of implicit surfaces reconstructed from point cloud data

2009, 29(07):  1901-1905. 

Asbtract ( )   PDF (941KB) ( )  

Related Articles | Metrics

In this paper, a novel rendering algorithm for implicit surfaces which are reconstructed from point cloud data by using a particle system was presented. First, an alternative initial technique based on bundles of parallel lines was used to find initial points that were evenly distributed on the surface. Because of its characteristics, the usual splitanddeath criterion of particle system was not needed. Second, each elliptical particle was moved towards a progressively lower energy state using a conjugate gradient method, which replaced the gradient-descent as an optimization method to avoid very long convergence time and irreconcilable oscillation around the minimum. Third, a greedy selection strategy was used to choose a subset of active particles which guarantee a hole-free approximation. Finally, a relaxation process further improves the curvature driven anisotropic particle sampling. The proposed elliptical particles were specially designed for splatbased representation and could be directly converted into elliptical surface Splats without any modification, thus high-performance rendering of complex implicit surfaces can be obtained.

Dynamic morphological simulation of tumor growth based on cellular automata

2009, 29(07):  1906-1909. 

Asbtract ( )   PDF (634KB) ( )  

Related Articles | Metrics

To simulate tumor deformation and the process of mechanical interaction between tumor and tissues, the authors have made a research into three aspects related to tumor growth.Cellular Automata (CA) was employed to simulate the growth process, and the cells states were updated after deformation to enhance the continuity of growth. The accuracy of penetrated depth will make great influence on deformation rationality and punctuality, and a method based on implicit surface was employed to calculate the penetrated depth vector. To increase the reality of tumor model, the authors also used implicit surface to render tumor model. The simulation results show that a better continuity of shape changes with state updating method of vertices, that the implicit surface calculated with the extracted vertices presents a smooth geometrical model and better reality effects, and that the improved strategy for penetrated depth vector achieves consistent depth vectors.

Low-cost 3D human reconstruction from multi-image

2009, 29(07):  1910-1913. 

Asbtract ( )   PDF (744KB) ( )  

Related Articles | Metrics

A method for low-cost 3D specific human model reconstruction from multi-image is proposed. First, a series of feature points are matched on images and generic model respectively. Then a deformation from radial basic function is performed to obtain specific model. After that, three kinds of filters are used to refine the specific model. Finally the surface of model is texture mapped by viewpoint-independent algorithm, and it is skin-skeletoned by vertex blending algorithm as well. The running experiment results show that this method is accurate, easy to implement and low-cost.

Fast template matching algorithm based on boundary cross correlation and image integration

2009, 29(07):  1914-1917. 

Asbtract ( )   PDF (567KB) ( )  

Related Articles | Metrics

A fast template matching algorithm was presented based on the theory of normalize cross correlation (NCC). With the goal of accelerating the matching speed and yielding exactly the same result as other matching methods，it is applied to calculate the similarity of bounded cross correlation and image integration. In this process, NCC must calculate cross correlation and selfcorrelation at each point. Self-correlation was calculated only first and then Holder inequality was applied. With the given threshold, it can remove the unsatisfied point, and reduce the complex computation of cross correlation, significantly accelerate template matching. The advantage of image integration is in the process of matching the integration of sub-image can be obtained quickly by the integration of the whole image calculated before matching. The proposed algorithm has been applied in the wire-bonder image recognition system, and the experiment results show that template matching algorithm is of fast speed and high accuracy and this algorithm has practical value.

Image segmentation approach of combining fuzzy clustering and graph cuts

2009, 29(07):  1918-1920. 

Asbtract ( )   PDF (512KB) ( )  

Related Articles | Metrics

This paper proposed an image segmentation method combining fuzzy Cmeans clustering and graph cuts. Based on graph cuts, considering the pixel space, establish a global energy function concerning labels, and make the FCM clustering center as terminal to found a multi-terminal network, get the global minimum of the energy function or the approximate minimum of the label function by the α expansion moving algorithm .Redemarcate all pixels between all classes and get the right targets. Experimental results show that the method has great improvement in accuracy, performance and noise immunity.

Surfel decimation method for point models

2009, 29(07):  1921-1923. 

Asbtract ( )   PDF (508KB) ( )  

Related Articles | Metrics

To efficiently simplify the densely sampled point model, a point sample data reduction method based on surfel reconstruction was presented. The surfel radius was generated and surfel decimation coefficient based on curve was made. According to surfel decimation coefficient, surfels were combined and a new surfel was generated based on given error. The point model was returned when the error reached given value or the surfel coefficient reached given value. Experimental results show that the algorithm can effectively simplify the dense sampled point models, and the decimation ratio is higher than other point decimation algorithm under control of the same error; the simplified model can preserve the original geometric shapes very well and prevent surface inanition.

Adaptive mouth region detection algorithm based on multi-color spaces

2009, 29(07):  1924-1926. 

Asbtract ( )   PDF (617KB) ( )  

Related Articles | Metrics

Mouth region detection is the prerequisite of lipreading. In this paper, an adaptive mouth region detection algorithm was proposed based on multicolor spaces. The algorithm combined color edge detection in RGB color space and threshold segmentation in HSV color space. Furthermore, according to the mouth position of the face, an adaptive lip localization method was introduced to detect the position of mouth baseline automatically. Then the rectangular area of mouth region was detected by projection method. Experimental results show that the proposed algorithm can locate mouth region fast, accurately and robustly.

Face recognition based on 2DLDA and SVM

2009, 29(07):  1927-1929. 

Asbtract ( )   PDF (423KB) ( )  

Related Articles | Metrics

"Small sample size" problem of LDA algorithm can be overcome by twodimensional LDA (2DLDA), and Support Vector Machine (SVM) has the characteristic of structural risk minimization. In this paper, two methods were combined and used for face recognition. Firstly, the original images were decomposed into highfrequency and lowfrequency components by Wavelet Transform (WT). The high-frequency components were ignored, while the low-frequency components can be obtained. Then, the liner discriminant features were extracted by 2DLDA, and "one vs rest", strategy of SVMs for multiclass classification was chosen to perform face recognition. Experimental results based on ORL (Olivetti Research Laboratory) face database and Yale face database show the validity of 2DLDA+SVM algorithm for face recognition.

Gray-level value natural logarithm ratio neighborhood filters method for image denoising

2009, 29(07):  1930-1932. 

Asbtract ( )   PDF (547KB) ( )  

Related Articles | Metrics

A new method based on gray-level value natural logarithm ratio Neighborhood filters is presented for image denoising in this work. Based on the present Neighborhood filters model，the principles, and the existing problem of the gray-level value were analyzed for the normalization Factor. As a solution, a new gray-level value natural logarithm ratio range filter kernel is pointed out for the Bilateral filtering, a representative method of Neighborhood filters. The new gray-level value natural logarithm ratio Bilateral Neighborhood filtering method can not only well restrain noise but also keep much more weak edges and details of an image, and preserve the original color transition of color images. Experimental results show that the effectiveness for image denoising with our method is better.

External parameters calibration algorithm for digital camera using circular target

2009, 29(07):  1933-1936. 

Asbtract ( )   PDF (527KB) ( )  

Related Articles | Metrics

External parameters calibration for digital camera is a key technology for field measurement and image processing while using computer vision. An external parameters calibration algorithm for digital camera using circular target was presented. Firstly, iterative learning approximation algorithm was designed for getting the coordinates of circle images center on the target. On this basis, external parameters of digital camera could be calculated using known intrinsic parameters of digital camera and imaging geometric model. The experimental results indicate that the presented algorithms in this paper have high accuracy and good stability.

Concentric circle detection based on chord mid-point Hough transform

2009, 29(07):  1937-1939. 

Asbtract ( )   PDF (406KB) ( )  

Related Articles | Metrics

This paper presented a Hough transform-based detection method of concentric circles. The core idea of the method is to reduce the Hough cumulative dimension of the circle detection based on classical Hough transform, and convert the threedimensional statistical parameter based on the traditional circle Hough transform to two-dimensional parametric statistics about the center of a circle and one-dimensional statistical parameter about the radius of a circle. According to the fact that a round of all the points must be inscribed circle round, the black and white points of the edge image were extracted, and a twodimensional array accumulator was established. Accumulator counts after scanning Pixeledge image with the accordance of distribution, and then the calculation of the center position was carried out. Finally, cumulate the radius of each circle center; determine the radius of concentric circles set. The whole process as opposed to the traditional Hough transform detection greatly reduces the computational complexity and demand for resources.

CT-PET image registration in the lung using parallel computing technology and hierarchical B-splines

2009, 29(07):  1940-1942. 

Asbtract ( )   PDF (501KB) ( )  

Related Articles | Metrics

In order to improve the efficiency of medical image registration, an image registration method based on hierarchical B-spline and parallel computing was proposed in this paper. First, landmarks were selected from the evenly distributed inner points and the outer contours of the images, and the correspondence points were automatically determined by using the parallel computing technology. Then the registration deformation was accomplished by hierarchical B-splines algorithm. Experimental results demonstrate that the method is efficient.

Denoising salt and pepper noise algorithm based on image statistical information

2009, 29(07):  1943-1946. 

Asbtract ( )   PDF (680KB) ( )  

Related Articles | Metrics

In this paper, a denoising algorithm with statistical information of image is proposed. The mean and variance of the central pixel and its neighborhood is used to cancel the salt and pepper noise of image. First, the author introduced the principles of the algorithm. Second, the median denoising algorithm, the adaptive median denoising algorithm and the denoising algorithm of this paper are all used to denoise the same image wirh the pollution of salt and pepper noise, also their comparison and analysis is displayed. Last, the author calculates and analyzes the complexity of the algorithm of this paper, and compares it with the complexity of the median denoising algorithm and the complexity of the adaptive median denoising algorithm, and concludes the reasonableness of the algorithm of this paper.

Image annotation and retrieval algorithm based on relevance feedback

2009, 29(07):  1947-1950. 

Asbtract ( )   PDF (601KB) ( )  

Related Articles | Metrics

To improve the performance of image annotation and retrieval, an image annotation and retrieval algorithm was proposed based on region segmenting and relevance feedback. The proposed algorithm used the relevance of visual characteristics and annotation information, and obtained a group of similar photos by clustering based on visual characteristics of regions. Then it calculated the similarities between the region and the nearest three classes, and merged the keyword probability vector (KPV) to get the most appropriate KPV. The proposed algorithm also adopted the users feedback information to adjust the weights between query words and each category to improve the accuracy of image retrieval. Simulation results prove that the proposed algorithm greatly improves precision and recall of image retrieval.

CIC decimation filter based algorithm of harmonic analysis

2009, 29(07):  1951-1953. 

Asbtract ( )   PDF (599KB) ( )  

Related Articles | Metrics

A CIC decimation filter based algorithm of harmonic analysis was proposed in order to eliminate the spectrum leakage generated by asynchronous sampling and obtain a higher accuracy of power system harmonic analysis. In the case of the preceding ADC oversampling, the integral cycles of signals were truncated by backward searching from the original asynchronous data, and the sampling frequency was synchronized with the fundamental frequency of the signal by CIC decimation filter based on frequency conversion, then the spectrum obtained after Fast Fourier Transform (FFT), and at last the amplitudes and phases of the harmonic components were calculated. Computer simulation and related calculations show that the proposed algorithm, compared with the traditional methods, reaches a higher accuracy. The proposed algorithm, as an easytouse and effective method, only needs a single cycle sampling points for harmonic measurement of fluctuant periodic signals.

Improved distributed algorithm of reducing energy consumption in data aggregation

2009, 29(07):  1952-1954. 

Asbtract ( )   PDF (462KB) ( )  

Related Articles | Metrics

Through studying the algorithm of serial data aggregation, it is found that its difficult in a real scenario to save the global routing knowledge for all sensors. In order to reduce the additional energy consumption, the authors proposed a new distributed algorithm to aggregate data. Multiple sensor nodes initiate the process of data aggregation simultaneously, transmiting information to cluster head through multichannel approach. All sensors choose the next transmission direction by themselves, avoiding saving the global knowledge. Simulation results show that this algorithm can reduce the transmission delay and energy consumption effectively.

Study on dynamic allocated message scheduling method

2009, 29(07):  1954-1956. 

Asbtract ( )   PDF (627KB) ( )  

Related Articles | Metrics

A dynamic distributed message scheduling method suitable for CAN bus networks was proposed in this paper. It could guarantee that all messages of the CAN network were transmitted to destination nodes successfully within given time, and could fairly allocate network bandwidth to message streams. In this method, the identifier of message frames was reconfigured, and its binary value was related to the lifetime of messages and the waiting time of messages and so on, the longer the waiting time of message or the shorter lifetime was, the higher priority level of message was. The necessary condition of full schedule ability of messages was also analyzed in this paper, and some experimental results verify the necessary condition is correct and the scheduling method was valid.

Dynamic Web service composition model based on domain ontology

2009, 29(07):  1957-1959. 

Asbtract ( )   PDF (461KB) ( )  

Related Articles | Metrics

A formal description of domain ontology and dynamic Web service composition was presented. Based on the composition of dynamic service process and service interface match method, a new domain ontologybased model was proposed. The model decomposed the service request into service processes first, then, based on the service interface match, it further decomposed every broken down service process into composition services. An example of customer subscription was also illustrated to show how to create and apply this new model in enterprise applications.

ARM and simulated annealing algorithm based music CAI system

2009, 29(07):  1960-1962. 

Asbtract ( )   PDF (409KB) ( )  

Related Articles | Metrics

As is known to all, it is difficult to give a reasonable and quantized advice to music learners. An ARM and simulated annealing algorithm based music Computer-Aided Interviewing (CAI) system was designed. The performance of this system applied to music-teaching shows its validity and superiority.

Parameters setting parameters and simulation of transient state model in engine thermal management based on KULI

2009, 29(07):  1963-1965. 

Asbtract ( )   PDF (613KB) ( )  

Related Articles | Metrics

On a basis of a heavy-duty diesel engine the cooling system model was established using KULI software, and the temperature of coolant and oil was simulated at transient state. The paper mainly described that how to set engine model parameters in a transient state model with KULI software. The results show that using the engine model, the simulated results are in good accordance with experimental data, so the model is reliable and the principle of parameters setting provides a method for setting engine thermal management transient model which can be used to other engines. The engine model supplies appropriate thermal boundary environment for Engine Thermal Management System (ETMS), and it can be used to simulate the whole ETMS.

Research on seamless resource discovery architecture based on ASP pattern of manufacturing grid

2009, 29(07):  1966-1969. 

Asbtract ( )   PDF (722KB) ( )  

Related Articles | Metrics

Resource sharing on manufacturing grid (MG) consists of two parts: resource supply and resource application, which are connected by resource discovery. From the perspective of resource discovery, composite structure of resource discovery based on Application Service Provider (ASP) pattern of MG was proposed from two levels: first, on architecture of MG, one structure based on a general enterprise service center was proposed; second, on deployment architecture of ASP, architecture of multilevel and tree overlay network was proposed. They support with each other and provide support of seamless structure for resource discovery. This structure can realize lamination retrieval and multispot resource discovery in broad area and has features such as being enterpriseoriented and applicationoriented and so on. Based on the architecture above, one parallel algorithm of resource discovery was designed whose performance was analyzed and compared with other algorithm. Finally, validity of this architecture is powerfully illustrated with a qualified example and realization about some parts of the architecture.

Model for print process planning based on polychromatic sets theory

2009, 29(07):  1970-1973. 

Asbtract ( )   PDF (595KB) ( )  

Related Articles | Metrics

Based on the study of printing process requirements and processing sequence, the computer aided process planning was adopted in printing production system, the polychromatic sets theory was applied to construct the mathematical model of printing process planning, and an algorithm to search the printing process path was designed based on this model. The model was comprised of a contour comprising matrix and a polychromatic graph. The contour comprising matrix was used to describe the capability of a printing production system. The polychromatic graph was used to describe the interaction among working procedures. The proposed algorithm can search the printing process path according to the machining requirements of printed matter, and was implemented by computer easily. Finally an example was analyzed to verify the proposed model and the algorithm.

Design and implementation of in-memory WebGIS based on Linux

2009, 29(07):  1974-1977. 

Asbtract ( )   PDF (656KB) ( )  

Related Articles | Metrics

From the perspective of computer storage systems, a new idea to speed up WebGIS response was presented, which used memory as data storage instead of normal hard disk to solve the I/O bottleneck problem of hard disk. To solve the volatile problem of RAM and develop inmemory WebGIS, a solution based on tmpfs and initrd was offered after analyzing three types of ramdisk techniques(ram disk, ramfs, tmpfs) in Linux and the composition of Linux operation system. The system composition, framework and memory usage strategy of the inmemory WebGIS were detailed. On this basis, based on 64bit Debian GNU/Linux and MapServer WebGIS platform, the implementation of an inMemory WebGIS application was described in six aspects：buliding WebGIS application on hard disk, building inmemory WebGIS application mirror image file, reconstructing initrd, planning memory, loading system and updating WebGIS application. The test result shows that using memory storage is an effective and proper way to speed up the response of a WebGIS application.

Voice interaction system model based on semantic search

2009, 29(07):  1978-1980. 

Asbtract ( )   PDF (618KB) ( )  

Related Articles | Metrics

In this paper the speech recognition based on semantic search is discussed, The difference between text search and semantic search is introduced. The Distributed speech recognition model was constructed. The ontology of speech recognition based on semantic search was established. Our study has shown that study of alternated system model based on semantic search is useful.

Domain learning-based ontology mapping method

2009, 29(07):  1981-1984. 

Asbtract ( )   PDF (742KB) ( )  

Related Articles | Metrics

Ontology mapping is the key problem of semantic data integration, discovering these mapping manually is inaccurate and time-consuming, using tools to automate discover these mapping between source ontology and target ontology is necessary. In this paper we propose a domain learning-based ontology mapping method which can discover mapping between the concepts of ontology, use domain knowledge and historical mappings to find the complex mapping rules about the ontology being matched, so they can be matched better. The experiments show the correctness and efficiency of our solution.

Design of inquiries terminal for non motorvehicles based on embedded technology

2009, 29(07):  1985-1987. 

Asbtract ( )   PDF (472KB) ( )  

Related Articles | Metrics

In order to enhance the efficiency of inquiring traffic in non motorvehicle management, inquiries terminal for non motorvehicles was designed by taking embedded technology as framework and integrating a new generation of electronic non motorvehicle license based on Radio Frequency Identification (RFID) technology, which has several functions such as RFID information acquisition, data matching in real-time, output display and so on. The hardware structure and software design of this system was described, and the embedded operation system μC/OS-Ⅱwas transplanted. In the end, realization of software functions based on multi-tasks and analysis of performance were stated.

Measurement study of PPLive IPTV system

2009, 29(07):  1988-1991. 

Asbtract ( )   PDF (579KB) ( )  

Related Articles | Metrics

The author had conducted systematic research to PPLive, proposed the PPLive crawler design proposal, and carried out related active measurement using this crawler as the platform to the PPLive network. Finally correlation analysis of the PPLive users behavior characteristics was given. The result provides an effective method for measurement study of the PPLive or other P2P IPTV systems.

Research of migrating workflow service method supporting exception handling

2009, 29(07):  1992-1995. 

Asbtract ( )   PDF (628KB) ( )  

Related Articles | Metrics

Migrating workflow is a mobile computing model in which the migrating instance executes tasks locally by making use of the work place services. How to ensure the liable workflow services provided by work place for the migrating instance is a key problem, which must be encountered in the design of the migrating workflow system. The paper proposed a workflow service method based on the case reasoning, which utilized service cases to keep the migrating instance working correctly and exception cases to support the tasks exception handling. Experimental case shows the method can strengthen the system's ability to handle the workflow service exceptions.

Greenhouse control system based on Fuzzy PID control

2009, 29(07):  1996-1999. 

Asbtract ( )   PDF (603KB) ( )  

Related Articles | Metrics

It is difficult for greenhouse to control temperature since it is usually of such characteristics as big inertia and big lag. Therefore, Fuzzy PID control was introduced into the system. Fuzzy PID control algorithm was used to design temperature control system according to the season of a year and the section of a day. By simulation and comparison with Fuzzy PID and general control， this paper concludes that the use of the method in the system can greatly improve the robustness and self-adaptability of the system. It can save energy and improve economic benefit.

Accomplishment and comparison of access control methods in enterprise search

2009, 29(07):  2000-2002. 

Asbtract ( )   PDF (455KB) ( )  

Related Articles | Metrics

Analyze and compare the features and key technologies of four popular access control methods in enterprise search. Test and compare the impact of search performance caused by these access control methods. Advanced three typical scenarioes for enterprise search application, and propose appropriate access control methods for them.

Collation-based judgment of modern Tibetan syllable

2009, 29(07):  2003-2005. 

Asbtract ( )   PDF (657KB) ( )  

Related Articles | Metrics

The concrete structure deciding of a Tibetan syllable is one of the fundamentals of Tibetan character collation. This paper established several standards that could judge the structure of a modern Tibetan syllable, and group those letter composition that failed to meet the standards as Tibetan Transliterating syllable. The first step was to decide whether a Tibetan character series meet the requirement of modern Tibetan syllable structure，according to the numbers and relative position of spacing and nonspacing letters in the composition. The second step was to judge whether every letter in the structure could meet the modern Tibetan orthography. In a short，the paper could judge whether a Tibetan letter composition is a modern Tibetan syllable, and if it is，the relative position of every letter could be welldecided. At last, the study selected Tibetan character series to verify the availability of the proposed algorithm.

Information processing with ascertaining Uyghur vocabulary and labeling marks set

2009, 29(07):  2006-2008. 

Asbtract ( )   PDF (545KB) ( )  

Related Articles | Metrics

This paper presented the research development of information processing with Uyghur set of tags marking progress. Firstly, this paper discussed that it is necessary to research and develop on "tagdimensional language marked"; secondly, the survey stated that it is limited for "tagdimensional language marked" depending on grammar terminology. And then the study introduced the latest results of English, Chineselevel analysis at home and abroad. According to the urgency of the classification system of Uighur semantic, the paper researched on "Uighur tagging tag collection" work ideas, formulation of the principles set tag and grammar semantic vocabulary of firstdegree tag sets, and gave the future development prospects.

Uyghur continuous speech recognition system based on HMM

2009, 29(07):  2009-2011. 

Asbtract ( )   PDF (689KB) ( )  

Related Articles | Metrics

Uyghur language is an agglutinative language. It is possible to produce a very large number of words from the same root with suffixes, so that the speech recognition of Uyghur language is very difficult. Combined with the characteristics of Uyghur language, this paper built a Uyghur continuous speech database, and designed the Hidden Markov Model (HMM) based Uyghur continuous speech recognition system by using the HTK (HMMToolKit). On the acoustic level, this paper selected triphone as the basic recognition unit, and used many methods such as decision tree, tiedstate triphones, fixing the silence models, increasing Gaussian mixture distribution to improve the precision of the models. On the language level, this study used the statisticsbased bigram language model. Finally, this paper presented some recognition experiments by using that system.

Chinese Mongolian numerals alignment based on Arabic numerals transformation

2009, 29(07):  2012-2015. 

Asbtract ( )   PDF (632KB) ( )  

Related Articles | Metrics

In the context of dictionary based Chinese Mongolian word alignment task, this paper proposed an approach to align Chinese Mongolian numerals. The main idea of the approach is to transform Chinese and Mongolian numerals into the Arabic numerals respectively before calculating their similarities and aligning them. The transformation is based on a two dimensional list called numtable, which includes the basic Chinese (or Mongolian) numerals and the corresponding Arabic numerals. Each numeral in the numtable is assigned an indicative value of "1" or "2", indicating whether the numeral in question is a multiple of "10". According to the indicative value, the transformation module carried out a series of inference and transformed the Chinese and Mongolian numerals into the Arabic numerals respectively. In the whole transforming process, not only the divergence between Chinese and Mongolian numerals were taken into account, but the different types of numerals in each language were handled differently.

Tibetan coding method and mutual conversion

2009, 29(07):  2016-2018. 

Asbtract ( )   PDF (464KB) ( )  

Related Articles | Metrics

This paper introduced the main Tibetan coding methods at home and abroad, and described mutual conversion from nonstandard to national standard（GB18030）,national standard to international standard according to different cording structure, and then realized it on Vista platform. It provides convenient tool for information exchange and processing in Tibetan and lays a foundation for the further development of Tibetan information.

Design of Tibetan segmentation dictionary and its algorithm study

2009, 29(07):  2019-2021. 

Asbtract ( )   PDF (445KB) ( )  

Related Articles | Metrics

With the development and improvement of natural language information processing, enormous linguistic material text processing has become a hot topic in the area of computational linguistics. One important reason is that it being able to collect the demanding knowledge from the huge corpus. However, the linguistic material text processing is based on grammar information dictionary. Combining the stanadard of Tibetan corpus segmenting tag, this paper discussed the establishment and design of Tibetan corpus segmentation and tags on Tibetan grammar information dictionary, and emphatically discussed the establishment of contexts, tags of grammar information, index structure and the means of calculation gathering.

Chinese-Uyghur machine translation system for phrase-based statistical translation

2009, 29(07):  2022-2025. 

Asbtract ( )   PDF (665KB) ( )  

Related Articles | Metrics

This paper described a ChineseUyghur machine translation system for phrasebased statistical translation. First the ChineseUyghur corpus was used to train the language model and the translation model. After using these models to decode the source statement, the authors got the best translation statement. The core decoding algorithm was the beam search algorithm, and the LatinScript Uyghur was used as the Uyghur corpus. The experimental results show that phrasebased statistical machine translation method can be used quickly and effectively to build a ChineseUygher machine translation platform.

Statistical analyzed on Tibetan shaped structure

2009, 29(07):  2029-2031. 

Asbtract ( )   PDF (425KB) ( )  

Related Articles | Metrics

Based on statistics from the Tibetan dictionary, calculated Modern Tibetan word the frequency of use in the Tibetan language. Analyzed the Tibetan at the level of a small character parts and get the characteristics of the word configuration. Meanwhile based on the structure of the Tibetan word of the initial consonant and the Final made a statistic. Finally , giving a number of characteristics of the Tibetan word from the way of the initial and the Final, which provides some basic data for the futural Tibetan studies and information processing.

Acoustic analysis of prosodic features of trisyllabic words in Uyghur language

2009, 29(07):  2032-2034. 

Asbtract ( )   PDF (400KB) ( )  

Related Articles | Metrics

In order to improve the naturalness of speech synthesis system, the present study is the first systematic empirical work on word stress in Uyghur language, the purpose of which is to establish a basis for the study of the prosody in Uyghur language. The experimental analysis was based on an acoustical database which contained prosodic measurements from 333 trisyllabic words, recorded by 1 male and 1 female speakers.The major part of the investigation concentrated on an acoustical analysis of the prosodic features of the trisyllabic Words in Uyghur language. The word stress for own words in Uyghur language is fixed on the last syllable. We believe that the study of this project not only has high research value for the study of Uyghur language but also has high research value for the study of entire Altay language family.