Applied Mechanics and Materials Vols. 713-715

Abstract: Rhythm annotation is the basement of speech analysis and speech synthesis technology for Tibetan. Base on the study of Chinese rhythm, and Tibetan characters, this paper has studied the rhythm annotation for the speech synthesis technology of Tibetan, and has designed a set of rhythm annotation rules, which consists of Latin transliteration, tone type, syllable type, stress type, and break indices. Getting the annotative texts by Praat; then extracting prosodic parameters by Matlab, which is including the parameter designing, parameter extracting and results saving three parts.

Abstract: The inductance and the capacitance on DC side are two important parameter of SVG. In this paper, the effect of inductance and capacitance on DC sides’ voltage control method of SVG is addressed.

Abstract: Due to the capacity constraint on traffic flow in multi-aerodrome terminal airspace, an algorithm to allocate traffic flow according to capacity in multi-aerodrome terminal airspace is studied by exploring collaborative decision making and law of economics. First, considering factors such as airline’s gross income, operation cost and delay cost, a traffic optimization model to maximize total airline’s profit is established, and constraints including capacity of multi-aerodrome terminal airspace are imposed as well. Then, an improved genetic algorithm to obtain optimal solutions to the optimization model is developed. Traffic flow can be better allocated according to capacity in multi-aerodrome terminal airspace by using the proposed strategy, while airline’s profit is increased simultaneously.

Abstract: Current prediction models for network traffic cannot accurately depict the multi-properties of the Internet traffic. This paper proposes a wavelet-based hybrid model prediction method for network traffic called CLWT model and proposes a prediction method for traffic based on this model. The traffic time series can be rapidly decomposed respectively into approximate time series and detail time series with LF and HF response. The approximate time series predicts by making use of Least Squares Support Vector Machine and proceeds error calibration by using Generalized Recurrent Nerve Network. The detail time series predict it by making use of self-adaption chaotic prediction methods after the medium-soft threshold noise reduction. Finally the prediction value of time series is got by making use of promoting wavelet reconstitution. The effectiveness for the prediction methods mentioned in the paper has been validated by simulation experiment. High prediction accuracy is obtained compared with the existing methods.

Abstract: An extraction method based on invariance geometric feature is proposed in this paper. This method extracts two types of feature from the object in an image. One type is five invariance statistical features of edge distance. The other is two invariance shape features: rectangular similarity feature and circular similarity feature. Moreover, this proposed method is used to extract defect features for steel plate surface. Its performance is tested in scale and rotation invariance and defects classification. Experimental results show that the novel geometric features have the ability of invariance and can improve the accuracy of classification.

Abstract: Aiming at the characteristics of the UAV camera, camera data nowadays, a new improved method is proposed based on putting the low-resolution video reconstruction into high-resolution video. First, the low-resolution video frame is done spectrum analysis by Fourier transform. Second, find the maximum gradient descent point to determine the cut off frequency. Finally making use of high-resolution images with high frequency detail, then motion compensated. Through POCS algorithm, then iterated, obtaining super-resolution reconstruction video and realizing the above by MATLAB simulation.

Abstract: Under the background of huge amounts of data in large-scale power grid, the active power optimization calculation is easy to fall into local optimal solution, and meanwhile the calculation demands a higher processing speed. Aiming at these questions, the farmer fishing algorithm which is applied to solve the problem of optimal distribution of active load for coal-fired power units is used to improve the cloud adaptive genetic algorithm (CAGA) for speeding up the convergence phase of CAGA. The concept of cloud computing algorithm is introduced, and parallel design has been done through MapReduce graphs. This method speeds up the calculation and improves the effectiveness of the active load optimization allocation calculation.

Abstract: Differential Evolution Algorithm (DE) is fast and stable, but it’s easy to fall into the local optimal solution and the population diversity reduces fast in the later period. In order to improve the algorithm optimization and convergence capability, this paper proposes an improved DE algorithm based on the new crossover strategy (CMDE). As to the Crossover-factor is decided by the proportion of the variance and the evolution process in each generation, so it can follow the process of evolution and constantly change; the added operation of Second Mutation can improve the capacity of solving problem, which algorithm falls into the local solution easily. With four standard test functions, the results show that the CMDE algorithm is superior to DE in convergence speed, precise and stability of algorithm.

Abstract: Mean shift algorithm is a robust approach toward feature space analysis, which has been wildly used for natural scene image and medical image segmentation. Due to fuzzy boundary and low accuracy of Mean shift segmentation method, this paper puts forward to an improved Mean shift segmentation method of high-resolution remote sensing image based on LBP and Canny features. The results show that this improved Mean shift segmentation access can enhance segmentation accuracy compared to the traditional Mean shift.

Abstract: Based on the principle of the image enhancement, various image enhancement methods are introduced, analyzed and studied. Because image enhancement is closely related to the property of the interested target, the habit of observers and the specific processing goal, the image enhancement is only aimed at the given process goal, too. According to different images, these image enhancement methods are simulated by the MATLAB tools. Through comparing the test results, the results show that different methods will give different effects. Without a common image enhancement method is suitable for various occasions.