Applied Mechanics and Materials Vols. 427-429

Abstract: In this paper, we present a coordinate rotation digital computer (CORDIC) based fast algorithm for power of two point DCT. The proposed algorithm has some distinguish advantages, such as regular data flow like the Cooley-Tukey FFT, identical post-scaling factor, and the rotation angles of the CORDICs in DCT are arithmetic sequence. By using the sum formula and double angle formula, we dramatically reduce the CORDIC types in the proposed algorithm. For the purpose of gain fast speed, we use Carry Save Adder (CSA) as the basic cell of each part and Carry Look-ahead Adder (CLA) to produce the outputs in the final level. Compared with other known architectures, the proposed 2-point DCT architecture has lower hardware complexity, higher throughput and better synchronization

Abstract: Wear particles recognition is a key link in the process of Ferrography analysis. Different kinds of wear particles vary greatly in texture, texture feature is one of the most important feature in wear particles recognition. Local Binary Pattern (LBP) is an efficient operator for texture description. The binary sequence of traditional LBP operator is obtained by the comparison between the gray value of the neighborhood and the gray value of the center pixel of the neighborhood, the comparison is too simple to cause the loss of the texture. In this paper, an improved LBP operator is presented for texture feature extraction and it is applied to the recognition of severe sliding particles, fatigue spall particles and laminar particles. The experimental results show that our method is an effective feature extraction method and obtains better recognition accuracy compared with other methods.

Abstract: Sense disambiguation is an important problem in pattern recognition. In this paper, a new algorithm of sense disambiguation is proposed, in which part-of-speech tags of the left word and the right word around the ambiguous word are extracted as discriminative features. At the same time, the bayesian model is selected as the sense disambiguation classifier and it is built based on discriminative features. The architecture of sense classification is given. The new algorithm is trained on sense-annotated corpus. Then it is used to determine its sense category. Experimental results show that the accuracy rate of disambiguation arrives at 60%.

Abstract: Optimal path finding algorithm is a very important research direction in computer science. It is a node search method based on graph theory. This paper focus on the fundamental researches of the optimal path finding algorithm. By comparing various path finding algorithms, this paper implement the A* algorithm and present a variant strategy of A* which is named Turning Point A*(TPA*). The experiment shows that searching with TPA* can speed up A* by an order of magnitude and more and report significant improvement.

Abstract: This paper is concerned with a TOPSIS method for fuzzy multiple attribute decision making, in which the information about attribute weights is completely known and the attribute values take form of intuitionistic fuzzy numbers. A class of distance for describing the deviation degrees between intuitionistic fuzzy sets is used to measure difference between two alternatives. A model of TOPSIS is designed with the introduction of the particular closeness coefficient composed of similarity degrees. Then, we apply the TOPSIS method to aggregate the fuzzy information corresponding to each alternative, and rank the alternatives according to their closeness coefficients. Finally, a numerical example is given to show the feasibility and effectiveness of the method.

Abstract: In view of the grid has a large number of harmonics, serious interference power quality, but the existing harmonic detection has many disadvantages, this paper uses genetic algorithm to optimize the BP neural network (GA-BP),Combined with fanaticism, independent component technique in the separation technology, achieve rapid separation and accurate fitting for harmonic reduction. The method to solve the problem of slow convergence speed of BP neural network and easier to fall into local optimal solution; Separation, independent component analysis in the signal could not accurate reduction the source signal. In this paper a signal containing harmonic component are simulated. Results show that GA-BP neural network optimization independent component analysis has convergence speed, high precision, compared with the single BP neural network and the independent component analysis of the superiority.

Abstract: It focuses Isomap isometric embedding algorithm is proposed to improve supervised isometric embedding algorithm (SIsomap). Both supervised manifold learning algorithm, using the introduction of adjustable parameters in the form of classes in the classification problem for the effective use of information, making the manifold learning algorithms for classification classification problems have a stronger effect. Finally, through a series of experiments to fully illustrate the proposed improvement of the effectiveness of the algorithm, the proposed oversight of the manifold learning algorithm can more effectively enhance manifold learning algorithms for classification problems

Abstract: It focuses on locally linear embedding algorithm into LLE proposed supervised locally linear embedding algorithm (SLLE). That supervised manifold learning algorithm, which introduced adjustable parameters to effectively use the classification information, so as to make the SLLE have a stronger effect for classification problems. Finally, through a series of experiments to fully illustrate the proposed improvement of the effectiveness of the algorithm, the proposed oversight of the manifold learning algorithm can more effectively enhance manifold learning algorithms for classification problems proficiency.

Abstract: A system for detecting and evaluating drivers gaze behavior was proposed. A system for recognizing the drivers unsafe gaze behavior was established using multi-level information and fusion decision method as well. The driving environment and condition is complex as well as the gaze behavior characteristics, and given that, a solution consists of patten classification and the multi-information decision-level fusion were put forward to estimate the different kind model of the driver's gaze behavior. In order to test the proposed strategies,the real time driver's gaze behavior detection system was established. The T characteristic curve proposed through the abnormal behavior parameters of the transverse width between the eyes and the vertical distance between mouth and the midpoint of two eyes, combined with the driver's eyelid closure and the proportion and location characteristics of iris - sclera were studied to get the characterization of the drivers gaze status information. The simulation results indicate that the adaptability and accuracy as well as the intelligent level is significantly improved by using the pattern classification and decision-making technology through multi-source information fusion.

Abstract: Association rules analysis is an important subject in data mining. At present, association rules mining algorithms frequently generate a large number of association rules, but most of the algorithm evaluations make advances only from an aspect, which makes the users select difficultly. Therefore, the comprehensive evaluation of association rules has become highly necessary. A comprehensive evaluation system of association rules based on the AHP (Analytic Hierarchy Process) was presented, which can evaluate the association rules from multi-angle and multi-dimensional. Many evaluation results are integrated into the system, eventually presenting a unified comprehensive coefficient to users. Practical data make it clear that the comprehensive evaluation system is rational and superior.