Applied Mechanics and Materials Vols. 303-306

Abstract: The local features based on interest point have achieved much success in action sensing recently. The interest point is not only limited to 2D space, but also extended to 3D space. We apply the 3D interest point to action sensing. A classic method to use 3D interest point is through creating a feature using histogram vector based on bag of words; some better methods take advantage of the position of each interest point besides the local feature; however, it’s difficult to position these points due to the complexity of an action. We propose a simple method to position each interest point, and create a new feature for action sensing. Evaluation of the approach on two sets of videos suggests its effectiveness.

Abstract: In this paper, data mining methods are used in the blast furnace production control . firstly introduces some main clustering algorithms.Then clustering analysis of blast furnace operation parameters is carried out by K-Means Clustering. Analysis and comparison of practical data is conducted to determine the optimal cluster number of this algorithm for blast furnace parameters analysis, and thus yield the ideal operating value for the parameters. The optimal threshold for blast furnace parameters is determined through statistical analysis, repeated experiments and field assessment, and the difference between blast furnace state as estimated and the practical one analyzed. Finally, the factor analysis method to reduce the dimension of parameters. Mining test in data of tangshan iron&steel shows that the method is effective in practical application.

Abstract: This paper proposes a novel approach for structure modal parameter identification. The approach changes the identification problem to an optimal one. The global optimal solutions for the required parameters, including frequency, damping ratio, amplitude and phase of the structure can be obtained by taking the advantage of the Partical Swarm Optimization. The results of numerical simulation studies showed that the accuracy of this method is comparatively higher, and the adjacent modal coupling has no effect on its accuracy. The FIR lowpass has effect on the phase’s identification.

Abstract: Kernel entropy component analysis (KECA) reveals the original data’s structure by kernel matrix. This structure is related to the Renyi entropy of the data. KECA maintains the invariance of the original data’s structure by keeping the data’s Renyi entropy unchanged. This paper described the original data by several components on the purpose of dimension reduction. Then the KECA was applied in celestial spectra reduction and was compared with Principal Component Analysis (PCA) and Kernel Principal Component Analysis (KPCA) by experiments. Experimental results show that the KECA is a good method in high-dimensional data reduction.

Abstract: Image segmentation methods that exploit multiscale information about images to be estimated have been extensively studied, typically using the Hidden Markov Tree (HMT) framework. we incorporate wavelet coefficients information of the original image in the form of Hidden Markov Tree model prior for the object segmentation. In this paper, we derive a generalized closed form inference scheme to exact determine the posterior likelihood at each iteration with definite number of iteration steps. Extensive experiments show that this method performs better than many competitive multiscale image segmentation methods.

Abstract: Abstract. Image segmentation is a key technology in image engineering, it occupy an important position. This paper introduces the watershed transform to Image of monolithic circuit processing method, and then introduced the watershed transform to Image of monolithic circuit segmentation and sample. The results show that, by using the watershed algorithm and morphological processing function, which is connected with a plurality of object images are segmented into a plurality of single object, to achieve the image segmentation, and as far as possible to reduce or eliminate the phenomenon of over-segmentation. Finally it points out the further direction of research.

Abstract: The analysis of the time sequence can be two ways in the time domain and frequency domain. But many financial time series exhibit strong non-stationary and long memory, which makes many traditional individually focused on the research and analysis of the time domain or frequency domain method is no longer applicable. In this paper, wavelet analysis and support vector machines for use in the time domain and frequency domain have the ability to characterize the local signal characteristics, location and mutation of the singular points and irregular mutation analysis, these mutations detected the degree of significance.

Abstract: Rough set theory is a kind of ambiguity and imprecision new mathematical tools, using precise mathematical analysis of imprecise system an ideal method. Rough set theory has powerful data reduction capability, this paper rough set theory to model the stock time series data, reduction, rule extraction, study the ups and downs of the relationship between the stock price, the use of advanced data mining techniques to dig out price linkage between stock association rules, has a very important significance.

Abstract: Results of autocorrelation analysis algorithm by the LabVIEW are different from the theoretical results. To address the problem, a modification of autocorrelation analysis is proposed in this paper. In the proposed approach, the circular correlation and linear compensation are employed to solve the distortion problem in the original algorithm. Simulation results show that the method can reduce errors of autocorrelation analysis effectively.

Abstract: Topology is one of the mechanisms to describe relationships between spatial objects and it is the basis for many spatial operations. The paper gives a survey of current main three dimensional topological data structures. Three dimensional topological data structures can be divided into manifold data structure and non-manifold data structure. Manifold data structure includes Winged-edge data structure, Half-edge data structure, Quad-edge data structure and so on. Non-manifold data structure includes facet-edge data structure, radial edge data structure and so on. The paper gives on overview of fundamental principles of these data structure. On this basis, advantages and disadvantages of these models are compared from more aspects. Through this research, we can provide theoretical basis and technical support for 3D building modeling, 3D cadastre modeling and other 3D fields.