Considering the lack of shape character in reconstruction of cross-cut English documents,this paper proposed an automatic reconstruction method of document fragment based on gray-level matrix.The method used eigenvectors of fragmented images to eliminat the negative effects of peer letters at different heights,extracted its features location as a dynamic clustering feature classification criteria,which provided text line characteristics for dynamic clustering and category sorting.This paper designed a neighborhood image stitching algorithm based on fragments of the original boundary matrix,one by one efficient re-modeling.Simulation results show that the method has a high degree of accuracy and is convenient to operate.It has good practical significance for rule fragments mosaics restored.

微博作为国内主流社交网站,信息量与日俱增.目前微博用户兴趣挖掘方法大多停留在研究用户浏览网页时点击行为、用户所发微博内容或所在社区等表象层面,尚未深入到微博用户使用特性层面.从用户微博内容出发,结合用户关注对象微博,提出一种改进作者主题模型UF_AT(users focus-author topic).最后对真实数据进行实验得出,模型在用户兴趣主题以及主题词概率值上均高于AT模型,而且用户兴趣主题准确、全面,同时验证了UF_AT模型在挖掘用户兴趣中的有效性.

As micro-blog grows more popular and widely used, micro-blogging site such as Sina has become a huge source of information, although the traditional method of texts, topic mining has been extensively applied research. For this special kind of text of micro-blogging, traditional text mining algorithm can not be well studied. In order to compensate the deficiencies of current topic mining for micro-blogging platform and considering the sparsity and multidimensional characteristics of micro-blogging, this paper proposes targeted pretreatment method and combines the users’micro-blogging data with AT model, then mining the micro-blog topics by gibbs sampling, getting users’interest through extracting the topics of authors. Through the experiments on a real data sets, as well as comparison with LDA models prove that the model can get micro-blog topics effectively.

Clustering result of k-means clustering algorithm is highly dependent on the choice of the initial cluster center. With regards to this, a text clustering algorithm based on improved Particle Swarm Optimization ( PSO ) is presented. Features of particle swarm algorithm and k-means algorithm are analysed. Considering the disadvantages of PSO including low solving precisions, high possibilities of being trapped in local optimization and premature convergence,self-regulating mechanism of inertia weight and cloud mutation operator are designed to improve PSO. Self-regulating mechanism of inertia weight adjusts the inertia weight dynamically according to the degree of the population evolution. Cloud mutation operator is based on stable tendency and randomness property of cloud model. The global best individual is used to complete mutation on particles. Those two algorithms are combined by taking advantages of power global search ability of PSO and strong capacity of local search of k-means. A particle is a group of clustering centers,and a sum of scatter within class is fitness function. Experimental results show that this algorithm is an accurate,efficient and stable text clustering algorithm.

Considering the problem including slow convergence rates, low solving precisions and easy to trap in local optimum of Particle Swarm Optimization(PSO) algorithm, a novel dual population hybrid algorithm named CGA-PSO is presented, which is based on Cloud Genetic Algorithm(CGA) and PSO algorithm. In this algorithm, the whole population is divided into two equal populations. CGA and PSO with self-adjusting inertia weight strategy are used in the process of evolution of two populations. Two populations share the best individual and eliminate the worst individual by exchanging information between the two groups of offspring as well as offspring and parent to complete the evolution, and a timely cloud mutation operation is given on poor fitness of individuals. Cloud mutation operation is based on stable tendency and randomness property of cloud model. The global best position and the global worst position are used to complete mutation on the part of the particle’s position. By testing five classical functions and comparing CGA-PSO with CGA, PSO and their optimization algorithms, the results show that the proposed algorithm has higher search efficiency, accuracy and rapid convergence speed, and stronger robustness.

A new family of object model,i.e.declarative family of object model,was imposed in this paper.The model was defined,geometric and topological structure of this model was presented,semantic of feature was specified by constraints,constraint graph was used to represent features in the model,grammar specification of model was provided.Validation of the model was verified in a instance,and insufficient of traditional procedural-based modeling was overcome,design efficiency of CAD modeling was developed,and design cost was reduced.

A new solving approach for constraint problem was proposed in this paper, the constraint problem needed to solve was decomposed not into single sub-problems, but into three types of sub-problems, namely, rigid subset, scalable subset and radial subset, and each type of subset corresponds a cluster of constraint problem. Based on cluster rewriting rule approach, a small set of rewriting rules were applied in constraint system, and then an incremental algorithm was applied, the generic solution will be get when there is no available rewriting rule to be applied. By this approach, we can determine that constraint system is well-constrained, under-constrained or over-constrained. The results reveal that the proposed method can efficiently process constraint problem.

This paper proposed a new technology of constraint resolving based on parameterization of freeform feature model. In this method,parameterized freeform feature definition point,established geometry constraint graph,decomposed constraint problem into triangle constraint and tetrahedron constraint,and solved each of them,constructed sub-problem as global result by using handedness rule. This arithmetic was implemented in the HUST-CAIDS,and the result model can meet designer requirement.

Aimed at the problems that the standard part library does not support heterogeneous CAD systems and the incompletion of the data information,a standard part library resource sharing framework based on cellular ontology was proposed.The cellular ontology model used Web Ontology Language (OWL) to develop the cellular ontology model.Using semantic mapping between legacy ontology and cellular ontology built the uniform representation of product model data,exchanged data information according to the cell,shielded the heterogeneity of information format,which implemented share of standard part library and real-time exchange of product model data among heterogeneous CAD systems.The applications in the synchronized collaborative design between Pro/E,UG and CATIA were also introduced to prove the feasibility of the theories above.

A novel algorithm of tracking topological changes of family of object models was imposed,relations between parameters and topology of models were established,critical values of parameters were computed,dependent entities of models were determined,stable intervals and parametric ranges were computed,and at last topological changes of family of object models were tracked accurately in this system.The algorithm was used in our HUST-CAID system,and intelligent of system was improved,parametric ranges was determined for designers.

To find effective and intuitive ways of modeling with freeform feature,freeform feature and freeform feature definition points are parameterized in this paper,a new method of creating freeform feature with wrapping parameters and constraints is presented.Mapping from numerical parameters to geometry is defined in the definition of freeform feature.Feature definition points are picked up on a trajectory,and a cross-section is built on each points,then skinning and sweeping operation are applied on these cross-section,as a result freeform feature fitting seamlessly to the target surface is built.

将多 Agent 技术引入到计算机支持的协同设计(CSCD)领域中来,有效地解决了设计者之间信息的共享、通信和协作问题.分析了基于多Agent的协同设计技术的特点,提出了一个基于多Agent的协同设计计算模型,利用这个计算模型和多Agent的关键技术,实理了一个基于多A-gent技术的复杂产品协同设计系统框架原型.

Solution strategy about feed formulation was analyzed,using multi-agent coordinating strategy and linear programming strategy,Agent model was constructed,optimization algorithm about feed formulation based on multi-agent coordinating strategy was proposed,and its correctness and feasibility was proved based on theory model about linear programming of multi-agent coordinating solution which was proposed by Bruin.The results of experiment indicated that this algorithm possessed solution ability well.