Advanced Materials Research Vols. 532-533

Abstract: For the actual results of visual perception for virtual soldier, the model of visual perception for virtual soldier based on limited rationality is established. The paper analyses the conditions which affect the visual perception for virtual soldier including geometrical gene, visual gene, moving gene and contrastive gene. The visibility of virtual soldier is computed and the persuasion for visual perception of virtual soldier is received.

Abstract: This paper presents a new multi-resolution Markov random field model in Contourlet domain for unsupervised texture image segmentation. In order to make full use of the merits of Contourlet transformation, we introduce the taditional MRMRF model into Contourlet domain, in a manner of variable interation between two components in the tradtional MRMRF model. Using this method, the new model can automatically estimate model parameters and produce accurate unsupervised segmentation results. The results obtained on synthetic texture images and remote sensing images demonstrate that a better segmentation is achieved by our model than the traditional MRMRF model.

Abstract: Data mining techniques give us a feasible method to deal with great amount of data, which is generated during the software developing. Many methods have been used in data mining, Bayesian networks become a focus currently. It is a powerful tool and can be used to do uncertain inference. Bayesian networks have several advantages for data modeling. This paper mainly discusses the definition and building of Bayesian networks, research software engineer based on data mining, and builder a application model of data mining in software engineer, description detail the core arithmetic of Bayesian network. An instance is presented to indicate the applications of Bayesian network in data mining technology final. The experimentation proves its practicality and availability.

Abstract: In order to solve the Gaussian kernel density-based background modeling, we propose a background modeling method based on an 8-neighborhood pixels sample set. In this method, we use the target pixel and its surrounding 8-neighborhood pixels to analyze whether it is included in the background or in the foreground sample space. Experimental results show that the method can judge the object as the background interference caused by the cyclical movement. By practical verification, the algorithm of the background modeling can full meet the requirements of the target detection algorithm.

Abstract: In this paper, it is analyzed the content and achieving mechanism of WM algorithm on account of frequently asked questions during the monitoring and filtering of Short Message Service(SMS) spam, and brought forward a kind of Chinese message with multi- patterns fuzzy matching method to improve WM algorithm to apply to Chinese spam message filtering. The method firstly does pretreatment of key words and fuzzification of monitor for message on account of expressive diversity of Chinese message, and then realizes initial matching by improving WM algorithm. Because the preliminary result is not the final one, it needs further fuzzily match for the result. It does particular introduction of the method and the model of monitor and filtration for Chinese SMS spam in the paper. The method’s correctness and performing efficiency also has been done experimental analysis, test and comparison experiment with former system that had not been done pretreatment. Experimental results show the validity of the improving algorithms. The method is testing further during the practice.

Abstract: This paper proves multicast management belongs to problems of HDP, provides the cost analytical model of multicast logical key tree which applies HDPT. Based on the above analysis, the author proposes an improved multicast logical key management scheme termed as L-R logical key tree management scheme. L-R logical key tree management scheme is a kind of imbalanced multicast logical key tree, which is based on periodic re-key, according member nodes’ joining and leaving different probability, it employs different multicast key management strategy. The paper provides the design and calculation of the scheme, and realizes it in a simulation. By comparing this scheme with the typical LKH one, it shows clearly the scheme proposed in the paper is of high effect and low cost.

Abstract: Wavelet transform is widely used and has good effect on image denoising. Wavelet transform has unique advantages in dealing with the smooth area of image but is not so perfect in high frequency areas such as the edges. However, curvelet transform can supply this gap when dealing with the high frequency areas because of the characteristic of anisotropy. In this paper, we proposed a new method which is based on the combination of wavelet transform and curvelet transform. Firstly, we detected the edges of the noisy-image using wavelet transform. Based on the edges we divided the image into two parts: the smoothness and the edges. Then, we used different transform methods to dispose different areas of the image, wavelet threshold denoising is used in smoothness while FDCT denoising is used in edges. Experimental results showed that we could get better visual effect and higher PSNR, which indicated that the proposed method is better than using wavelet transform or curvelet transform respectively.

Abstract: We have used the metaphor of ant colonies to define "the immune system", a class of distributed algorithms for combinatorial optimization. In this paper we analyze some properties of Ant-cycle, the up to now best performing of the ant algorithms we have tested. 4.2% Re alloy 1320 °C solution treatment can significantly reduce the elements in the interdendritic / arm region segregation, after the high temperature air-cooling solution, γ'-phase combination of butterfly present form, after aging at 1100 °C, the size of about 0.45 m to cubic γ'-phase coherent manner mosaic in g matrix, aging temperature increased, γ' phase size too grew up.

Abstract: The management of university web sites is becoming more critical than before due to the rapid growth of the population dependent on the world wide web as the most important (if not the only) information source. A university can spread its research outcomes and education achievements through its web site, and consequently gain visibility and influence from the web population. Webometrics Ranking of World Universities (WR) proposed by Centre for Scientific Information and Documentation (CINDOC-CSIC), which ranks the university web sites, has obtained much attention recently. The rankings of WR are well recognized as an important index for universities willing to promote themselves by the internet technology. In this paper, we proposed WRES as an early warning system for Webometrics Rankings. WRES gathers the WR indices from the WWW automatically in flexible periods, and provides useful information in real time for the managers of university web sites. If the WR ranking of an institution is below the expected position according to their academic performance, university authorities should reconsider their web policy, by promoting substantial increases of the volume and quality of their electronic publications. Besides, the web site manages may adopt effective approaches to promote their WR rankings according to the hints given by WRES.

Abstract: Many issues such as low accuracy rate, poor filtration efficiency, and too much network resource occupancy exist in most information filtering systems in the current market. Therefore, there is a great demand to develop a new network filtering system. In this paper, agent techniques are first reviewed, and then the architecture and the working mechanism of a multiagent-based information filtering system is presented. Finally, the design method for core components of the system is given.