Applied Mechanics and Materials Vols. 427-429

Abstract: This paper presents a facial expression recognition algorithm based on multi-channel integration of Gabor feature. First, a Gabor wavelet filter extracts facial features with 5 scales and 8 orientations, and then transform the 40 channels into 13 channels according to the maximum rule presented in this paper. Second, we reduce the dimension of expression features by the method of PCA+LDA. At last, expression features are classified using the nearest neighbor method. The experiments involve two databases and show that the proposed algorithm can recognize facial expression in high rate.

Abstract: From the request of practical applications,according to the basic theory of IIR filters, a scheme of hardware implementation is worked out combining with the fact that coefficients of numerator and denominator of transfer function are fixed and the structural feature of selected FPGA. From the clew of implementing stratified ,modularized and parameterized design ,the thesis describes the hardware implementation of the IIR filter with VHDL and schematic diagram design method.

Abstract: A novel imaging approach for dispersive target electromagnetic imaging based on the time-frequency analysis is presented in this paper. On the foundation of researching the time-domain imaging principle of dispersive target, the basic theory of back projection imaging algorithm and time-frequency analysis is discussed. The sampling data obtained by simulation of target electromagnetic wave scattering which is accomplished by finite-difference time domain method (FDTD). According to the simulating data, after disposing by Wigner-Ville transfer, the sampling data is processed integral management in frequency-domain, and the imaging experiment is carried out. The experiment results show that the improved imaging arithmetic could estimate the shape of dispersive target and improve the imaging quality.

Abstract: Stepped frequency signal is an important high distance resolution signal, this property is realized by combining several pulse signal, so it is easy to be effected by targets radial motion. Here we analyzed how targets radial motion affects the synthesis HRRP, then a method of velocity measurement is provided, then compensation using the predicted velocity.

Abstract: This paper proposed and implemented a general identification process of videos containing moving target cells, using mature digital image processing technology and MATLAB simulation tools. The processing steps included video sub-framing, image reading, image preprocessing, target recognition, and etc. For videos containing moving target cells, three groups of experiments were tested to verify the feasibility of the processing algorithm. Results found that the algorithm could accurately identify the targets in the videos. Thus, the presented processing algorithm is acceptable to be a general identification method for videos containing moving targets.

Abstract: It is essential to detect the gross errors for improving the precision of soft sensing model. Clustering technique was used to detect gross error in this paper. Based on Fuzzy C-Means clustering algorithm (FCM) and Differential Evolution (DE), the proposed algorithm can detect the gross errors in modeling data for a soft sensor. The numerical experiments result shows that the algorithm is effectively.

Abstract: Since wireless sensor networks consist of sensors with limited battery energy, a major design goal is to maximize the lifetime of sensor network. To improve measurement accuracy and prolong network lifetime, reducing data traffic is needed. In the clustering-based wireless sensor networks, a novel data aggregation algorithm based on OPT and Layida Method is proposed. In the proposed method, Layida Method preprocesses data and data fusion model for data integration are used. Its availability is proved by comparing with the results of two existing algorithms.

Abstract: This paper analyses time-domain waveform and its unilateral energy spectral density based on the Gaussian pulse former 12 derivative functions. We could conclude that pulse forming factor has a closer relationship with different order number derived function affected the energy spectral density. Therefore, in order to maximize the approximate Federal Communications Commission (FCC) emission mask as much as possible, an iterative algorithm is proposed to optimize the linear combination of the Gaussian pulse former 12 order derived function. Compared with single differential Gaussian pulse, the simulation results show that after linear combination, new ultra-wideband pulse signal could greatly meet the spectrum utilization and indoor radiation mask standards promulgated by the FCC.

Abstract: A new face image feature extraction and recognition algorithm based on Scale Invariant Feature Transform (SIFT) and Local Linary Patterns (LBP) is proposed in this paper. Firstly, a set of keypoints are extracted from images by using the SIFT algorithm; Secondly, each keypoint is described by LBP patterns; Finally, a combination of the global and local similarity is adopted to calculate the matching results for face images. Calculation results show that the algorithm can reduce the matching dimension of feature points, improve the recognition rate and perspective; it has nice robustness against the interferences such as rotation, lighting and expression.

Abstract: In order to solve the problem that empirical mode decomposition (EMD) will cause false components in the process of signal decomposition, a method of false component discriminant of EMD based on Kolmogorov-Smirnov test was put forward. First, the original signal was decomposed into several intrinsic mode functions (IMFs) by EMD. Then the K-S test was used to calculate the similarity between each IMF and the original signal. The reasonable similarity threshold was selected for judging the authenticity of the IMFs. The IMFs of which the similarity values were less than the threshold value were determined to be the false components. The others of which the similarity values were greater than the threshold value were determined to be the real components. As a result, the false components were removed and the real components were remained. The vibration signal of bearing experiment indicated that the method of K-S test could discriminate the real components and the false components obviously. Then the false components were removed quickly and accurately and the real components of the original signal were obtained.