Applied Mechanics and Materials Vols. 397-400

Abstract: By analyzing the gray level features of transition region, a new underwater image transition region extraction method based on Support Vector Machine (SVM) is presented. At first, a vector is constructed to fully describe the transition region, which includes local complexity, local difference and neighborhood homogeneity. Then, SVM is applied to train and classify the set of feature vectors, so that the transition region of the underwater image is extracted. Finally, the segmentation threshold is determined by mean of the histogram of the transition region, and the binary result was yielded. The experimental results show that the proposed algorithm can achieve a better transition region extraction and segmentation performance, and automatically select the optimal threshold for transition region extraction.

Abstract: For the continuously-varying dynamic pressure flutter test data in the high-speed wind tunnel; this paper analyzes the strengths and weaknesses of the data Processing Method in this field. The data selection and stacking mode is improved. A Butterworth IIR digital filter is used for digital filtering. The power spectral density function is computed by Fourier transform of the autocorrelation function. A pick-hold method is used to predict the flutter critical point. It is indicated that our method features high accuracy and efficiency by theoretical analysis and wind tunnel flutter test.

Abstract: The key parts of multi-function and all-electric rheometer (MAR) designed by the authors were introduced. The data processing method based on frequency-domain analysis were established. Intrinsic vibration displacement is defined to eliminate the effect of steady velocity displacement and extract the useful information of vibration displacement. The actual frequency, amplitude and phase of the superimposed vibration displacement and the entrance pressure were obtained by the frequency-domain analysis of the intrinsic vibration displacement and the entrance pressure. The control precision of the superimposed vibration displacement frequency and amplitude of MAR proved to be high. The direct current component of entrance pressure can be seen as the mean entrance pressure which stands for the flow resistance. The study of mean entrance pressure is useful in discovering the energy consumption mechanism of the polymer processing introduced vibration energy. Loss angle can be calculate according to the phase of entrance pressure and superimposed vibration displacement. The higher order harmonic of the entrance pressure can be used to characterize the nonlinear viscoelasticity of polymer melts. The factors which depend on the shear rate and influence the viscoelasitic properties can be studied by test the loss angle based on MAR.

Abstract: The P300 oddball task is the most popular paradigm in the existing BCI systems. Recently using auditory stimuli in P300 oddball task arises since it gives much freedom to the BCI user. In this paper, we present a novel BCI paradigm using P300 and P100 responses. Since P300 and P100 responses occur in the frontal lobe and the temporal lobe respectively, so that we can use these responses stimulated by an audio in a single task. The main advantage of our designed paradigm is that we can obtain two different kinds of responses in a single trial EEG task. In the EEG data analysis, we first employ the multivariate empirical mode decomposition (MEMD) algorithm to extract P300 and P100 components. And then, we employ a support vector machine (SVM) technique for the feature classification.

Abstract: The data processing of traffic volume is a key issue in the traffic planning. In order to solve the problem of data processing on traffic volume, the bi-level programming model built is employed to work out production-attraction value of each district and O-D distribution matrix. In this model, upper part is to calculate the minimized variance between investigated value and calculated value. While the lower part of this model aims at the fact that travelers turn to the shortest path. The upper model is solved by genetic algorithm. And the result from an example application reveals that we can obtain reasonable production-attraction value and O-D distribution matrix by this method.

Abstract: A method of the image acquisition based on digital signal processor (DSP) is introduced. DSP, complex programmable logic device (CPLD) and contact image sensor (CIS) are combined in the hardware design, and the time-sequence analysis of the image acquisition process is also presented. Practical application indicates that this method has high accuracy and is rapid enough to satisfy the requirement of real-time acquisition.

Abstract: According to the actual needs of underwater acoustic homing system in system debugging and experiment parameters setting, a new design of real-time water target simulator is introduced. This system simulates the acoustic reflection characteristics of targets, receives active detection signal and emits signals of reflection echo characteristics. The target simulator can response up to two active detection signals in the condition of interference. This system is complete, accurate and can be extended. Users can monitor the working state of simulator and change the parameters of simulator. As a result of an echoic tank experiment and lake experiment, the system has been successfully applied to the underwater homing system.

Abstract: Fundus images are complex images with more details, and on the basis of the inadequate fuzzy enhancement algorithm proposed by Pal et al, this article propose an improved algorithm of rough set for fundus image enhancement. The fundus image will be multi-scale decomposed by wavelet transform firstly, and then the subgraphs are enhanced by using rough set to improve the visual effects; finally the processed sub-images will be reconstructed and generated a new-enhanced image. Compared with the Pal algorithm, the new algorithm not only overcomes its weaknesses that the threshold is set with a fixed value, but also reduces the number of iterations. Experimental results show that the improved enhancement algorithm has a better effect on the fundus image enhancement, and the various details of fundus images can be shown better.

Abstract: In the estimation of time delay, there always would not appear obvious peak with the basic cross-correlation (CC). In order to solve the problem of the basic cross-correlation method, this essay represents an improved time delay estimation method based on the generalized cross-correlation (GCC) and combines with the microphone array structure to achieve sound source localization. Finally, the simulation results show that this method could measure the sound source’s location accurately with noise and reverberation, and the distance positioning error is less than 10cm, the direction angle error is below 3°.

Abstract: When matching pursuits (MP) method was used for noise suppression of ultrasonic testing signals, the number of matched atoms affects the denoising performance. The relationship between the number of the matched atoms and denoising capability was analyzed, using the root mean square error (RMSE) and improvement of signal-to-noise ratio (SNR) to evaluate denoising performance. The simulated signals with white noise at different SNR and experimental signal with white noise and grain noise were analyzed respectively, and the results show that the MP method can remove the white noise and grain noise effectively. Moreover the best denoising performance can be arrived if the number of matched atoms is appropriate. At last, the selection principle of atoms number is given.