Automatic damage detection of engineering ceramics ground surface based on texture analysis

Ground textures seriously interfere with the exact identification of grinding damage. The common nondestructive testing techniques for engineering ceramics are limited by their difficulty and cost. Therefore, this paper proposes a global image reconstruction scheme in ground texture surface using Fourier transform (FT). The lines associated with high-energy frequency components in the spectrum that represent ground texture information can be detected by Hough transform (HT), and the corresponding high-energy frequency components are set to zero. Then the spectrum image is back-transformed into the spatial domain image with inverse Fourier transform (IFT). In the reconstructed image, the main ground texture information has been removed, whereas the surface defects information is preserved. Finally, Canny edge detection is used to extract damage image in the reconstructed image. The experimental results of damage detection for the ground texture surfaces of engineering ceramics have shown that the proposed method is effective.     

Joint application of feature extraction based on EMD-AR strategy and multi-class classifier based on LS-SVM in EMG motion classification

This paper presents an effective and efficient combination of feature extraction and multi-class classifier for motion classification by analyzing the surface electromyografic(sEMG) signals. In contrast to the existing methods,considering the non-stationary and nonlinear characteristics of EMG signals,to get the more separable feature set,we introduce the empirical mode decomposition(EMD) to decompose the original EMG signals into several intrinsic mode functions(IMFs) and then compute the coefficients of autoregressive models of each IMF to form the feature set. Based on the least squares support vector machines(LS-SVMs) ,the multi-class classifier is designed and constructed to classify various motions. The results of contrastive experiments showed that the accuracy of motion recognition is improved with the described classification scheme. Furthermore,compared with other classifiers using different features,the excellent performance indicated the potential of the SVM techniques embedding the EMD-AR kernel in motion classification.     

Noise Reduction of Welding Defect Image Based on NSCT and Anisotropic Diffusion简

In order to reduce noise effectively in the welding defect image and preserve the minutiae information, a noise reduction method of welding defect image based on nonsubsampled contourlet transform （NSCT） and anisot-ropic diffusion is proposed. Firstly, an X-ray welding defect image is decomposed by NSCT. Then total variation （TV） model and Catte_PM model are used for the obtained low-pass component and band-pass components, respec-tively. Finally, the denoised image is synthesized by inverse NSCT. Experimental results show that, compared with the hybrid method of wavelet threshold shrinkage with TV diffusion, the method combining NSCT with P_Laplace diffu-sion, and the method combining contourlet with TV model and adaptive contrast diffusion, the proposed method has a great improvement in the aspects of subjective visual effect, peak signal-to-noise ratio （PSNR） and mean-square error （MSE）. Noise is suppressed more effectively and the minutiae information is preserved better in the image.     

Method and application of wavelet shrinkage denoising based on genetic algorithm

INTRODUCTION Most underwater acoustic signals received by sonar are corrupted inevitably by unpredictable noise sources. In most cases, sonar system is adversely influenced by noise so disturbing that target signals cannot be detected and classified correctly so that pre-processing is necessary to reduce the noise as much as possible. Methods of noise reduction based on wavelet transform have been developed extensively in pre-vious literatures (Xu et al., 1994; Sun, 1998), and include wav…     

Texture classification based on EMD and FFT

INTRODUCTION Multi-scale is one of the main features of natural images, a series of methods for representing the quality of images are presented, such as multi-scale technique based on diffusion equation (Perona and Malik, 1990), image pyramid (Burt and Adelson, 1983) and wavelet (Mallat, 1989). Bidimensional empirical mode decomposition (BEMD) (Nunes et al., 2005; Linderhed, 2004) is a new multi-scale analysis method proposed recently. The difference between BEMD and traditional mul…     

Experimental study of structural damage identification based on WPT and coupling NN

Too many sensors and data information in structural health monitoring system raise the problem of how to realize multi-sensor information fusion. An experiment on a three-story frame structure was conducted to obtain vibration test data in 36 damage cases. A coupling neural network (NN) based on multi-sensor information fusion is proposed to achieve identification of damage occurrence, damage localization and damage quantification, respectively. First, wavelet packet transform (WPT) is used to extract features of vibration test data from structure with different damage extent. Then, data fusion is conducted by assembling feature vectors of different type sensors. Finally, three sets of coupling NN are constructed to implement decision fusion and damage identification. The results of experimental study proved the validity and feasibility of the proposed methodology.     

减法聚类和粒子群优化TS模糊神经网络的驾驶疲劳融合检测(英文)

为提高基于单一特征检测算法的准确率和可靠性,提出基于多个特征的驾驶疲劳融合检测算法.从直接反映驾驶员疲劳的2个面部特征和间接反映疲劳的1个车辆行为特征2个方面对驾驶疲劳进行综合检测.该算法运用TS模糊神经网络来识别驾驶疲劳,采用减法聚类对网络进行结构辨识,确定模糊规则的条数及相关参数的初始值,并改进了粒子群优化算法对网络进行训练.仿真和实车实验表明,该算法不仅能有效改善TS模糊神经网络的收敛速度和识别精度,而且能提高驾驶疲劳的检测正确率.