Advanced Materials Research Vols. 756-759

Abstract: In order to overcome the issues that Support Vector Machine is sensitive to the outlier and noise points, Fuzzy Support Vector Machine (FSVM) is proposed. The key issue to solve the FSVM is determinate the fuzzy membership. This paper gives an overview of construction algorithm of the fuzzy membership. We also give an algorithm to solve FSVM that is derived from improved-SMO algorithm.

Abstract: Camera calibration is the first step of positioning using binocular vision. Owning to the approximation capability of the neural network, a complex mathematical model needed by traditional calibration methods can be avoided. However the general neural network methods have their drawbacks to reduce its accuracy. This paper presents searching algorithm for the best structure and parameters of a neural network using an improved genetic algorithm (GA). The experiments show that this method can be used to establish a mapping between 2D coordinates and 3D coordinates directly and accurately, which is better than traditional calibration and general BP network methods.

Abstract: A critical limitation for the current use of Unmanned Aerial Vehicles (UAVs) is represented by the lack of aerial refueling capabilities. This paper describes the results of an effort on the modeling of the refueling soft hose realized by FEM techniques and on the algorithms of the docking control. The algorithms of the docking maneuver is based on Machine Vision (MV).The simulation results with a vision technique capable of estimating the drogue position is presented. Simulation results suggest the proposed algorithm performs the same accuracy with extremely fast speed compared with the current employed algorithms.

Abstract: Usually, the algorithm of constructing cost-sensitive decision tree assume that all types of cost can be converted into a unified units of the same price, apparently how to construct an cost conversion function is an challenge. In this paper, a strategy of constructing heterogeneous cost-sensitive decision tree is designed and the different cost are take into account together in split attribute selection. Whats more, an attribute selection model based on heterogeneous cost-sensitive is constructed and the pruning strategy based on cost-sencitive is designed. The experimental results show that the proposed method is correct and more efficient than the present other methods

Abstract: TFIDF algorithm was used for feature weighting in text classification. But the result of classification was not very well because of lack of class information in feature weighting. The known class information in the training set was used to improve the traditional TFIDF feature weight algorithm. Class distinction ability and class description ability were introduced, respectively expressed by inverse class frequency and term frequency in class, document frequency in class. A new feature weight algorithm based on class information, TF_IDT, was proposed. Naïve Bayes classifier was used to test the algorithm. The precision, recall and F1 measure were significantly increased. Macro F1 measure raise by 6.46%. It was proved to be useful for improving text classification to use class information in feature weighting. In addition, the computational complexity of the proposed algorithm was lower and more suitable for use in fields of limited computing capability.

Abstract: The design and optimization of urban-rural dual-directions logistics network is a substantial important issue, which will directly affect the development of the urban-rural integration in China. A reasonable scheme of logistics network will contribute to supply efficient logistics services to customers scattering in urban and rural areas. In this paper, we consider a variant of the Location-Routing-Problem (LRP), namely the LRP with simultaneous pickup and delivery in specially background (LRPSB). The objective of LRPSB is to minimize the total system cost, including depot location cost and vehicle routing cost, and implement and control the effective dual-direct commodity flow to meet customers requirement by simultaneously locating the depots and designing the vehicle routes that satisfy pickup and delivery demand of customer at the same time. A nonlinear mixed integrated programming model is formulated for the problem. Since such integrated logistics network design problems belong to a class of NP-hard problems, we propose a two-phase heuristic approach based on Tabu Search, tp-TS, to solve the large size problem and an initialization procedure to generate an initial solution for the tp-TS. We then empirically evaluate the strengths of the proposed formulations with respect to their ability to find optimal solutions or strong lower bounds, and investigate the effectiveness of the proposed heuristic approach. Computational results show that the proposed heuristic approach is computationally efficient in finding good quality solutions for the LRPSB.

Abstract: Network structure such as blood vessels in medical images are important features for computer-aided diagnosis and follow-up of many diseases. In this study, a new model-based segmentation method is proposed to detect blood vessels in medical images. The Local Binary Fitting (LBF) model with statistical distribution function is used for this purpose. The brain tissues and cerebral vessels in the image are modeled by Gaussian distribution and uniform distribution respectively. The region distribution combined with the LBF model is used in curve evolution. And the level set method is developed to implement the curve evolution to assure high efficiency of the cerebrovascular segmentation. Comparisons with the LBF method show that our model can achieve better results.

Abstract: the mutual exclusion relationships among data items are reflected by negative association rules, whitch is very important on the decision-making analysis. In the last several years, negative association rules are frequently researched, while the study object of it is single mining of database now. With the development of database technology, multi-database mining is more and more important. On the basis of analyzing the related technology, research status and shortage of present negative association rules mining , the selecting rules, weighted synthesis and algorithm are discussed on multi-datobase.

Abstract: Under the background of information integration, the semantics of ontology integration is still an open issue. In this paper, we propose a revised distributed interpretation which is adapted from distributed description logics. In our proposal, ontology integration is taken as global ontology and local ontologies connected by ontology mapping. They are respectively interpreted with DL semantics and Semantic Import semantics. In this way, our method can facilitate understanding and maintenance of ontology integration.

Abstract: Photo-electricity target with low cost, high efficiency and high reliability is widely used for gunnery training in military affairs and athletic sports. The automatic target reporting system based on image processing is constructed and some experimental is done. The result is shown that compared to the way of manual target, automatic target reporting system has the character of easy to operate, low-cost and easy to popularize.