Advanced Materials Research Vols. 989-994

Abstract: Fractal technology can simulate the growth of trees very well.There have been many research on using fractal theory to model trees,but most methods have not further explore the tree’s natural phenomena such as swinging driven by wind Firstly, this paper gave a modeling method to generate the trees based on the three-dimensional fractal L system; Secondly, according to the hierarchy of the strength characteristics of the wind and trees progression, this paper also gave the wind model to simulate the natural wind farm.In the end , this paper made some simulation about the model of “Wind-Blowing & Tree-Swing and Leaves-Falling & Braches-Broken” .

Abstract: Software evolution process model (EPM) is a knowledge-intensive process which is described in EPDL(Software Evolution Process Description Language) and modelled by semi-formal approach based on EPMM(Software Evolution Process Meta-Model). EPM’s execution model (EEM) is represented by logic programming to create the knowledge base of EPM during constructing the EEM. Only its needing all kinds of resources are satisfied by system execution environment is activity in EEM implemented. The paper discussed the method and the algorithm of preparing resources is respectively presented in order to execute the EEM automatically. The converter program is realized and the correct results have presented to prove the correctness of the method.

Abstract: In this paper we construct model in computer forensics by using Benford’s Law, which is one of three famous laws in statistics in linguistics. Especially, we use Benford’s Law to analyze data in file analysis and Anomaly detection. We illustrate that Benford’s Law is able to supply the strategy for data analysis. The future work, Benford’s Law will be the theoretical rule for intelligent compression of mass data.

Abstract: In this paper we propose to use Zipf-like distribution to predict popularity data in storage systems. It can estimate prediction parameters according to the present statistics of I/O access. We classify the popularity data from every trace, and analyze the prediction rate through the classified popularity data’s characteristic. We synthesize the analysis results in different prediction time granularity and prediction popularity data queue. Finally, we use block I/O traces to discuss the effectiveness of prediction method. The discussion and analysis results indicate that this prediction method can predict the popularity data efficiently.

Abstract: This paper employs the Minimum Arborescence model to handle the spatial structure analysis of a mathematical formula, and transforms the directed graph to a trigeminal tree. Also, we utilizes the preorder traversal algorithm of the Ternary-Tree to generate MathML, and display the MathML with Web browser. At last, the experimental results illustrates the recognition rates of online handwritten mathematical formula in various cases.

Abstract: For the greedy algorithm required a large number of iterations, this paper proposed one adaptive bit allocation algorithm based on Hughes-Hartogs algorithm which was improved greedy algorithm .Under the constraint of bit error rate and data transmission rate, the algorithm first used the Chow algorithm to allocate some of the bits, and then used the greedy algorithm to allocate the remaining bits. When minimize the total power by this algorithm, the iterations of this algorithm were significantly less than the greedy algorithm. By computer simulation, the results showed that performance of this algorithm was very close to the greedy algorithm, and the number of this algorithm’s iterations was 7.4%~34% of the greedy algorithm.

Abstract: A 1450mm cold rolling mill roll stepped cooling control technology has been studied. The basic principle of stepped cooling control has been analyzed. The important role of stepped cooling control in the control of thermal crown has been elaborated. In order to analyze impact of stepped cooling on the sectional shape of the strip, the roll under the influence of the cooling water temperature field has been established. The model of optimal cooling capacity in each segment has been established. The model has been successfully applied to a 1450mm cold rolling mill roll, and a good effect has been achieved.

Abstract: K-Means algorithm is a one of the most famous unsupervised clustering algorithm. It has many disadvantages, such as sensitivity to the initial clustering centers and computes all the data points multiple times when facing the increasing data volume. In order to overcome the above limitations, this paper proposes to make use of density idea to find k cluster centers by adjusting the threshold. Finally, we design and implementation of the K-Means algorithm on the modern Graphic Processing Unit (GPU). The ratio of distance between classes to distance within classes and speedup are used as evaluation criteria. The experiments indicate that the proposed algorithm significantly improves the stability and efficiency of K-Means algorithm.

Abstract: The fast image processing based on iterative erosion principle can complete separation of colony image of adhesion target quickly. This algorithm is composed of local threshold method of image binarization; the morphological processing, analyzing; removing the edge based on Hough Transform and using improved iterative erosion algorithm to adhesion target segmentation and counting. The algorithm has scale invariance and self-adaption for recognition different size colony target. The conditional expansion algorithm designed in the paper can effectively modify the count mistakes which are caused by fragments formed by the segmentation of large size targets in the iterative and erosion processes. The algorithm has been applied in portable colony counting instrument on ARM11 embedded platform. Twenty heterotrophic bacterial samples have been tested in experiments. The results show that this algorithm can realize the image segmentation rapidly and effectively and the time of analysis is less than 2 seconds. The images edge detection rate can reach 100%. The detection accuracy can get the requirements of GB 4789.2-2010 and achieve deviation of 3% in colonies’ total number less than 300CFU.The applied of algorithm is widely in the similar particle image analysis system.

Abstract: Semantic retrieval method stands at the crossroads between Natural Language Processing and Machine Intelligent. This paper makes analysis on the semantic search method and research on concept similarity algorithm, and discusses the factor of weight’s influence on concept similarity as well. On this basis, this paper proposed a new semantic search method based on ontology, and apply it to the tourism information retrieval, which intellectualized tourism information retrieval service.