Applied Mechanics and Materials Vols. 644-650

Abstract: In order to reduce bit-rate and still maintain fine distortion performance,this paper proposed a method of LSF quantization based on speech unvoiced/voiced classification. This method use differential LSF parameters from unvoiced/voiced database to train codebook. And using this method can suppress the quantization error propagation caused by directly vector quantization of LSF parameters. Experimental results show that using this method to quantify LSF will have a better quality while allocating the same number of bits.

Abstract: Different regions in Foggy images have different contrast attenuations. This paper presented a defogging algorithm based on Statistical Region Merging. Firstly, we segmented the image by Mean Shift; Secondly, we merged these divided regions through Statistical Region Merging; Finally, we enhanced the contrast of different regions by Dynamic Segmentation Histogram Equalization contrast enhancement. The experimental results show that our defogging method is more effective than other methods.

Abstract: The safety evaluation of fishing vessels is a system and uncertain problem. A fishing vessel safety evaluation algorithm based on cloud model was proposed in this paper. Firstly, we established index system and weights by the AHP; Secondly, the cloud generator completed the conversion between qualitative and quantitative indicators; Thirdly, the evaluation cloud model was achieved by the comprehensive cloud algorithm; Finally, the forward cloud generator restored the droplet distribution. Compared with the fuzzy comprehensive evaluation method, the experimental results show that our work is effective and accurate.

Abstract: With the further reform and market division in the telecommunication industry, there are more and more choices for customers to select telecom products and operators, which lead to the fiercer competition for customers between telecom operators. As the technical method to identify customers churn, the data mining can help the telecom competitors to analyze some seemingly unrelated data into meaningful information. On the basis of the research on the vital problems in the telecom companies, this paper explains how to apply data mining techniques to customer churn analysis, proposes the specific procedures and technology solutions to prevent the customer churn and builds the models of the data mining by analyzing the related algorithm. Finally, based on the systematical analysis on theory and method to data mining, the paper draws the conclusion that the customers churn listing and tree algorithm can solve the practical problems of the customer churn in telecommunication industry.

Abstract: The essay is included as following: Standard Transfer Specification, DES algorithm, human program design, the characteristic of DES algorithm, the safety of DES algorithm and 3DES algorithm. Analyzing DES encryption and decryption algorithm results, we could well know the characteristic of DES algorithm. This algorithm is successfully tested on VC++ 6.0.

Abstract: By means of word-segmentation technology in TRIP database and each word that appears in a database will be account in detail, a kind of self-constructed category dictionary (SCC-dictionary) in Chinese text classification is proposed. For solving high dimension and sparseness problem exit in vector space model, a four-dimensional feature vector space model (FFVSM) is presented in this paper. With Support Vector Machine (SVM) algorithm, the text classifier is designed. Experimental results show there are two achievements in this paper: first, SCC-dictionary can replace the artificial-written dictionary with the same effect; second, the FFVSM will not only reduce the computing load than high-dimensional feature vector space model, but also keep the precision of classification as 86.87%, recall rate as 95.12%, and F1 value as 90.81%.

Abstract: Cutter suction dredgers perform a major part in the field of dredging engineering in harbors, fairways, and land reclamation. However, there are many parameters in cutter suction dredger operation so that it is difficult to guarantee the stability of production. In consideration of the issue of enormous parameters in dredging operation, mathematical dimensional reduction method which uses multivariate primary component analysis is proposed. The method can calculate the contribution rate and cumulative contribution rate of each parameter and then select the principal components which influents the production and energy consumption. These parameters represent the majority of the original data information, while not interrelated with each other. The primary components can be used to guide the regulation and control of the parameters, reduce regulatory parameters and operational complexity and provide a theoretical basis for intelligent automation of dredging operations.

Abstract: The algorithm in this paper is based on the basic principle of SVM. The sale data classification SVM classifier is designed using this algorithm. Also three classifiers including traditional grid search algorithm, ZGenetic Algorithm and Particle Swarm Optimization are used to do the comparison experiments of classification. Result shows that the improved grid-search algorithm can reduce the SVM classifier’s computational complexity effectively and improve its performance and classification accuracy.

Abstract: Additive Gauss white noise is one of the most commonly observed interferences in practical engineering applications. This paper proposed an algorithm for the adaptive determination of the optimal wavelet decomposition level based on Jarque-Bera test in efforts to solve the filtering problem of additive white Gaussian noise signal. By, The optimal decomposition level of wavelet is determined by testing the white noise which was realized by calculating skewness (S) and kurtosis (K) of the parameters. With signal-to-noise ratio (SNR) as the measurement index, simulation results show that the presented algorithm have higher accuracy, and better filtering effect on low SNR signals compared with nonparametric test methods.

Abstract: A deep study on the equivalent conditions between and implications. Several equivalent theorems were given, which help to better understand the relationships between them, also enrich the content of fuzzy implication theory.