ID	原文	译文
5714	首先,对信号进行小波分解,达到去噪目的;然后,将短时能量、短时平均幅值、过零率及频带能量值作为原始特征矢量,并结合Fisher判别准则进行特征选择,以此构造低维特征向量;	First of all, the signal wavelet decomposition, achieve denoising;Then short-term energy, short-term, zero crossing rate and average amplitude frequency band energy feature vectors as the original, based on feature selection, Fisher's discriminant criterion to construct low dimensional feature vector;
5715	最后,对两类声目标的实测样本数据进行特征提取,并采用支持向量机和K近邻两种分类器对该特征提取方法的有效性进行校验。	Finally, two kinds of feature extraction of the sample data of acoustic targets, and USES the support vector machine (SVM) and K neighbor classifier to check the effectiveness of the proposed feature extraction method.
5716	实验结果表明,采用"时域+频域+线性判别分析"的特征提取方法简单有效,且与单一时域或频域的特征提取方法相比,识别率更高。	Experimental results show that the "time domain, frequency domain and linear discriminant analysis feature extraction method is simple and effective, and with a single time domain or frequency domain feature extraction methods, compared to a higher recognition rate.
5717	现行的齿轮可靠性建模与分析方法往往只考虑了固有不确定性的影响,而忽略了实际中由于知识的缺乏而导致的认知不确定性对可靠性分析结果的影响。	Gear reliability modeling and analysis of the current methods tend to consider the influence of the inherent uncertainty, while ignoring the practice due to the lack of knowledge of cognitive uncertainty on the results of reliability analysis.
5718	本文给出了基于性能裕量的确信可靠度计算流程和算法,应用不确定理论描述了齿轮部分设计参数和影响因素的认知不确定性,定量地计算了接触疲劳和弯曲疲劳下的齿轮确信可靠度并进行了参数敏感度分析。	Are presented in this paper based on the performance of the process and the algorithm for calculating the credible degree that the allowance of sure, the uncertainty theory describes the gear parts design parameters and influence factors of cognitive uncertainty, quantitatively calculated the contact fatigue and bending fatigue of the gear that reliability and parameter sensitivity analysis was carried out.
5719	此外,将确信可靠度与未考虑参数认知不确定性的可靠度进行比较,进一步说明了考虑认知不确定性的必要性。	In addition, will sure reliability and not considering the reliability of cognitive uncertainty parameter comparison, consider the necessity of cognitive uncertainty was explained in further steps.
5720	主要研究光学与合成孔径雷达(synthetic aperture radar,SAR)图像的配准。	Main research optical and SAR (synthetic aperture radar, SAR) image registration.
5721	光学图像与SAR图像间完全不同的成像特性以及后者中斑点噪声的影响,增加了二者之间的配准难度。	Between the optical image and SAR image speckle noise in the different imaging features, and the latter, the effect of increasing the difficulty between the registration.
5722	此外,SAR图像数据量随着分辨率的升高而增大,这也对配准算法的实时性提出了更高的要求。	In addition, the SAR image data quantity increases with the rising of resolution, this is the registration algorithm of real time put forward higher requirements.
5723	针对这些问题,提出了一种逐步求精的光学与SAR图像配准方法。	Step by step to solve these problems, this paper proposes a refinement of optical and SAR image registration method.