Table of Content

    16 November 2008
    Volume 43 Issue 11
    Navigation method of the monocular vision based mobile robot
    DONG Ji-wen,YANG Sen,LU Shou-yin
    J4. 2008, 43(11):  1-04 .  doi:
    Abstract ( 861 )   PDF (427KB) ( 878 )   Save
    Related Articles | Metrics

    Due to the complex floor circumstances of the outdoors and the greatly changed intensity of light, the visual navigation of a mobile robot is still a problem. A new monocular robot's navigation method based on color image processing was proposed for solving this problem. By use of the image partition of the HSI model and mathematic morphology, the trajectory was picked out and the stop-sign drawn was recognized beforehand from the image. Then the “sampling estimated” method has used to calculate the parameters of navigation. According to the experiment this method is less impacted by light and floor circumstance, less error of aberrancy, better real-time and robustness.

    A diversity-guided two stages multi-objective particle swarm optimizer
    ZHENG Xiang-Wei,LIU Hong
    J4. 2008, 43(11):  5-10 .  doi:
    Abstract ( 896 )   PDF (605KB) ( 875 )   Save
    Related Articles | Metrics
    Multi-objective particle swarm optimizers are often trapped in local optima, converge slowly and cost more function evaluations. Therefore, a diversity-guided two-stage MOPSO (DTSPSO) was proposed. DTSPSO dynamically selects different mutation operators according to current population diversity and divides into two stages according to its ways of selecting leaders. In addition, Pareto dominance ranking and crowding distance were used to fix the size of the external archive. Experiments were carried out on several classical benchmark functions for multi-objective optimization problems and the results show that DTSPSO is effective in solving various multi-objective optimization problems.
    Variational models for image diffusion based on gradient and Laplacian
    PAN Zhen-kuan,WEI Wei-bo,ZHANG Hai-tao
    J4. 2008, 43(11):  11-16 .  doi:
    Abstract ( 1242 )   PDF (862KB) ( 785 )   Save
    Related Articles | Metrics
    Variational methods for image diffusion have been widely applied to image restoration with edge preserving or enhancement. The traditional models based only on gradients of image intensity can result in staircase effects beyond edges. So a hybrid variational image diffusion model using gradient and Laplacian was presented to reduce staircasing during image diffusion with edge preservation or enhancement. For the design of smooth terms in a diffusion model, forward and backward diffusion conditions for a one-dimension model based on first and second derivatives were first derived, which were extended to two-dimension image space. Some numerical results validated the model using a finite difference scheme.
    An implementing method of information management systems based on the business rule
    LIN Jin-jiao,PU Hai-tao
    J4. 2008, 43(11):  17-21 .  doi:
    Abstract ( 799 )   PDF (277KB) ( 668 )   Save
    Related Articles | Metrics
    Information systems should flexibly and quickly adapt to variable requirements. An information system implementation method was proposed that only modifying the business rule can achieve information systems adaptability. Based on a steady business pool consisting of business granules and business objects, variable factors on the business object aggregate business granule operates were separated from information systems codes and stored as the business rule, which was implemented through the corresponding business rule engines. Therefore, information systems based on this method can flexibly and quickly adapt to variable requirements through configuring the business rule. Experiments and examples were provided to evaluate the feasibility of the proposed method.
    Research on fuzzy multi-objective evaluation for a financing scheme of construction enterprise
    LIU Pei-de,ZHANG Xin
    J4. 2008, 43(11):  22-26 .  doi:
    Abstract ( 923 )   PDF (265KB) ( 745 )   Save
    Related Articles | Metrics
    A multi-objective comprehensive evaluation method was proposed based on a triangular fuzzy number, which is aimed at the evaluation and selection of a financing plan for a construction enterprise. First, an index system for evaluation of the construction enterprise’s financing flan was established based on related literature. Then, the evaluation values of weights for different indexes and alternatives in terms of every index, which is in the form of a linguistic term, were transformed into triangular fuzzy numbers, and subsequently the multi-objective evaluation model was constructed by stratified evaluation and group decision theory. Finally, an illustrative example was given to demonstrate the feasibility and easiness of the use of this proposed method.
    A hybrid evolutionary modeling algorithm for dynamic systems
    SONG Le-hui,CHENG Yue-gui,PAN Meng
    J4. 2008, 43(11):  27-30 .  doi:
    Abstract ( 878 )   PDF (720KB) ( 749 )   Save
    Related Articles | Metrics
    To overcome the difficulties and the draw backs in modeling dynamic systems by traditional methods, a hybrid evolutionary modeling algorithm (HEMA) was proposed to model the Dynamic system with a system of ordinary differential equations (ODEs) whose main idea was to embed string coded idea and evolution algorithm optimize system. The model structure was shown by string structure, and the evolution algorithm was used to optimize the structure and parameters of a model. It succeeded in automating the modeling process of systems of ODEs for dynamic systems. The result of experiments indicates that, by running the HEMA, the computer can search out multiple superior ODEs models within a reasonable time. Compared with the combination of a genetic algorithm (GA), which was employed to optimize the parameters of a model, and a genetic programming method, which was employed to optimize a model structure, this algorithm, possesses advantages, such as a more intelligent modeling process, more flexible and various model structures, higher precision of fitting and predicting data.
    Computer-aided design of a surface glazing wall based on fractals
    WANG Wen ,MA Xue-qiang,LIU Hong
    J4. 2008, 43(11):  31-35 .  doi:
    Abstract ( 1122 )   PDF (937KB) ( 1001 )   Save
    Related Articles | Metrics
    A new method of designing a glazing wall was developed based on fractal geometry and the surface equation. By means of this method, various fractal architectural images can be generated with a computer program in a short time. These images not only follow certain rules but are also random. They can expand the imagination of designers and stimulate their inspiration of creation. The method was written in VB computer programming language. The results show this method is effective.
    An OTC intelligently choosing system based on uncertainty reasoning and Fuzzy model
    LI Xin
    J4. 2008, 43(11):  36-39 .  doi:
    Abstract ( 853 )   PDF (548KB) ( 617 )   Save
    Related Articles | Metrics
    For the problems that the OTC used maybe not be proper for symptoms and it is difficult to choose the most satisfactory one from a lot of OTCs, an intelligently OTC chosen system was given. The system can diagnose common slight illnesses at an expert-level, gets all of the OTC which is proper for the diagnosed illness, and then according to the different aspects and different degrees of people focued on, it can evaluate the satisfaction with the OTC via two-level fuzzy synthetic judgement, and can recommend with satisfaction from high to low. This system can help people to correctly diagnose an illness and easily get the satisfactory OTC, and can instruct them to use the OTC safely and reasonably.
    An ant colony clustering algorithm on a manifold
    WANG Zong-li,LIU Xi-yu
    J4. 2008, 43(11):  40-43 .  doi:
    Abstract ( 918 )   PDF (435KB) ( 882 )   Save
    Related Articles | Metrics
    An entropy-based ant colony algorithm and a manifold learning method were introduced. The manifold learning method based on dynamical tangent space alignment was combined with an ant colony algorithm, giving a new idea of a clustering algorithm. This algorithm can improve the speed of convergence.
    Research on the indoors channel fading model in WSN
    ZHANG Xian-yi ,WANG Ying-long ,GUO Qiang,ZHAO Hong-lei
    J4. 2008, 43(11):  44-47 .  doi:
    Abstract ( 1061 )   PDF (621KB) ( 953 )   Save
    Related Articles | Metrics
    For the uncertainty of wireless sensor networks in an indoor physical environment, a model about channel fading was proposed. According to the theoretical approach in the wireless channel fading model and the analysis of the indoors and outdoors experimental data in the channel fading model coming from the testing system in wireless sensor networks, the various parameters of the channel fading model, were obtained and the indoor channel fading model in wireless sensor networks was confirmed. A practical experiment shows that this model is better than the rest of the wireless channel fading models in node location and deployment.
    The prediction model of financial distress of Chinese listed corporations based on a hybrid RPR model
    ZHU Shi-wei,SAI Ying
    J4. 2008, 43(11):  48-53 .  doi:
    Abstract ( 1017 )   PDF (324KB) ( 864 )   Save
    Related Articles | Metrics
    A new hybrid RPR model for predicting financial distress of Chinese listed corporations was put forth. The proposed model integrates the rough set(RS), the principle component analysis(PCA) method and the radial-basis function neural network(RBFN). Besides the traditional financial indicators, the cash-flow indicators were introduced, which can perfectly reflect the real-time financial situation of a corporation. In the model, the RS was applied to reduce the indicator, the PCA was employed to select indicators and reduce dimensions, the RBFN was used as a prediction tool for the corporate financial situation. Experimental results suggest that the model has high prediction accuracy and execution efficiency.
    An improved algorithm based on sets operation for mining frequent itemsets
    LOU Lan-fang,PAN Qing-xian
    J4. 2008, 43(11):  54-57 .  doi:
    Abstract ( 1033 )   PDF (238KB) ( 846 )   Save
    Related Articles | Metrics
    Mining association rules is an important issue in data mining and one of its key steps is generating frequent itemsets. A frequent itemsets mining algorithm based on sets operation was presented and compared with the classical algorithms apriori. This algorithm needs only to scan the database once. Experiments indicate the new algorithm is very efficient.
    An outlier-analysis algorithm based on the grid model
    YAN Zong-kui,SHI Bing
    J4. 2008, 43(11):  58-60 .  doi:
    Abstract ( 971 )   PDF (216KB) ( 748 )   Save
    Related Articles | Metrics

    To find the outlier in a data set more quickly and efficiently, an outlier-analysis based on the gird model was provided. This algorithm gives the way to partition the data space by the grid model, defines the boundary value of judging if there is an outlier existing in one grid, and gives the algorithm, which can be used in detecting the outliers correctly with less time.

    Research on P2P-based heterogeneous database data synchronization
    MENG Lei,SUN Yan-jie
    J4. 2008, 43(11):  61-66 .  doi:
    Abstract ( 794 )   PDF (796KB) ( 894 )   Save
    Related Articles | Metrics
    The existing problems in data integration and data synchronization were analyzed, especially the requirements of heterogeneous data synchronization and efficient transport. A heterogeneous database data synchronization model was given. By transferring into Extensible markup language files, the difference between heterogeneous data sources was shielded. A P2P-based optimal algorithm ensures real-time transports, which is essential to business applications. A data synchronization tool was developed and used in real applications. By comparing the system performance, it was shown that this tool can solve a heterogeneous problem and can keep high speed transmission when the network situation was not so good.
    An improved Apriori algorithm for mining association rules
    LIU Xing-tao,SHI Bing,XIE Ying-wen
    J4. 2008, 43(11):  67-71 .  doi:
    Abstract ( 1294 )   PDF (263KB) ( 823 )   Save
    Related Articles | Metrics
    For the bottlenecks of the Apriori algorithm, which restrict the efficiency of the Apriori algorithm, an optimized method was presented, which can take advantage of a two-dimensional array marker bit to achieve transaction reduction in association with taking advantage of order item to achieve item reduction. Reducing the times of joining as well as the number of scannings of the database will shorten the scan time. This algorithm takes advantage of order itemsets to improve the strategy, which is used to determine whether to join or not. And it removes useless transactions step by the step based on the transformation of a marker bit to reduce the number of transactions and items, while reducing the time of judgment. The results of an experiment show that the improved algorithm is more efficient.
    A retrieval model of an educational resource grid based-on hierarchical P2P
    CHEN Kun,LIU Fang-ai,XING Chang-ming
    J4. 2008, 43(11):  72-76 .  doi:
    Abstract ( 836 )   PDF (485KB) ( 858 )   Save
    Related Articles | Metrics
    Based on the educational resource grid, the organizational structure of the grid was discussed. In view of the hierarchical expansion tree structure of the educational resource grid, a distributed retrieval model of an educational resource grid was proposed based-on hierarchical P2P (DRM-HP2P), which can use Web Services technology to encapsulate the local resource retrieval of resource nodes, use the P2P way to carry on the intra-domain retrieval, forward retrieval requests to other domains through management node of this domain, then start several intradomain retrieval, and finally return the merged results. An application shows that this model has a high efficiency.
    SHI Kai-quan
    J4. 2008, 43(11):  77-84 .  doi:
    Abstract ( 807 )   PDF (289KB) ( 1133 )   Save
    Related Articles | Metrics
    By using Cantor’s sets, Packet sets were presented and called P-sets in brief. P-sets are composed of interior P-sets X and outer P-sets XF. The structure of P-sets and the relation between P-sets and Cantor’s sets were given. P-sets have dynamic characteristics. Under a static-dynamic environment, Cantor’s sets are the special case of P-sets, while P-sets were the general case of Cantor’s sets; and the interior P-decomposition theorem and outer P-decomposition theorem of P-sets were proposed. By using P-sets, integration knowledge, the structure of integration knowledge, and the applications of P-sets in dynamic knowledge discovery were given.
    Attribute dependence mining and recognition of knowledge
    LIU Ruo-hui,YAN Chuan-jiang,SHI Kai-quan
    J4. 2008, 43(11):  85-90 .  doi:
    Abstract ( 762 )   PDF (244KB) ( 563 )   Save
    Related Articles | Metrics
    If there are new attributes supplemented to α, then the elements in [x] will decrease; if some attributes in α were deleted, then the elements in [x]will increase. By using the above two characteristics, the concepts of ladder knowledge, the generation of ladder knowledge and the attribute dependence of knowledge were presented, the attribute dependence mining theorem of knowledge and the attribute dependence mining-state recognition criterion of knowledge were proposed, and the applications of attribute dependence mining of knowledge were also given.
    Rough area and double rough integrals
    ZHOU Yang ,L Kun ,
    J4. 2008, 43(11):  91-96 .  doi:
    Abstract ( 984 )   PDF (223KB) ( 784 )   Save
    Related Articles | Metrics
    Based on the functions generated from the upper approximation and the lower approximation in function one direction S-rough sets, the concept and characteristics of rough area and double rough integrals were given. In addition, the double rough integral was proved to be an ordinary integral and rough integral promotion. The double rough integral has two-direction dynamic characteristics, and it can provide an effective tool to resolve some practical problems such as output variation of energy.