语谱特征的身份认证向量识别方法

针对采用梅尔频率倒谱系数（mel-frequency cepstrum coefficient,MFCC）作为身份认证向量（identity vector,i-vector）进行说话人识别存在语音信息不全的问题,提出一种基于语谱特征的身份认证向量识别说话人的方法。语音信号经过预加重、分帧加窗预处理之后,通过短时傅立叶变换转换成语谱图,语谱图被提交到高斯通用背景模型,在高维均值超向量空间中选择合适的低维线性子空间流型结构以构造符合正态分布的向量作为身份认证向量。这些获取的身份认证向量经过线性判别性分析实现降维并存储。最后采用对数似然比（log-likelihood ratio,LLR）方法对训练和测试阶段的i-vector进行评分,完成说话人识别。以TIMIT数据库为标准的数值实验结果表明,相比采用MFCC作为特征的识别方法,研究的等错误率（equal error rate,EER）更低。     

A face recognition method based on modular PCA and SVD

In order to solve the problem that modular PCA method is sensitive to translation, rotation and other geometric transform, a face recognition method based on modular PCA and singular value decomposition (SVD) is proposed. The PCA features of sub image and SVD features are extracted respectively. The distance measure that fuses information of modular PCA and SVD is obtained. Minimum distance classifier is used to face recognition. Experimental results on ORL human face database show that the proposed method can obtain higher recognition rate.     

Performance analysis on improved PDA AI for moving target tracking

Aim at the problem that the EO imaging tracking system is inconsistent with the model of probabilistic data association with amplitude information (PDA AI), which supposes that the greater the amplitude value is, the greater the probability of being the tracked target will be, a modified PDA AI (MPDA AI) is presented . Based on the fact that the amplitude and the motion of the interested target are consistent in a short period, the MPDA AI models the amplitude information and the motion information of the target as well as their consistency with Markov stationary signal to analyze the association procedure of motion and amplitude. The lower bounds of Cramer Rao estimation error for PDA AI and MPDA AI are calculated and discussed in detail. The theoretical analysis and experimental results show that estimating the target motion with the MPDA AI will be more accurate and more reliable than estimating with the original PDA AI.     

Accurate measurement for objects space attitude based on planar target

On the basis of studying the accurately identification algorithm of planar target and transformation model of the space vector, the imaging target models and spatial attitude measurement models are established. Then, a new topological positioning method is put forward. Finally, the four-wheel location measurement is given as an example. The results demonstrate that the method has better anti-interference and higher measurement precision, which provides a feasible method for a machine vision measurement system by using the plane target.     

A contour extraction algorithm for digital images based on water balloons snake model

Due to weak edges and low contrast areas in digital images, such as a left ventricle in Magnetic Resonance Imaging (MRI) images, a deformation curve may leak from the outer boundary when images are segmented using the traditional snake model. An improved algorithm based on the water balloons snake model is introduced to solve this problem. Firstly, the barycenter and the boundary shape area of the left ventricle in a digital MRI image are obtained through mathematical morphology. The left ventricular watershed line is obtained next using a watershed transform algorithm, which is taken as the initial deformation curve in a balloon snake model to capture the contour. Different snake models are tested to digitally tagged MRI images of left ventricles of small pigs. Contrast results indicate that the water balloons snake model can solve the leakage problem in traditional snake and balloon snake models. In addition, the processing speed is improved significantly.     

模式识别的斜拉桥损伤诊断动力指纹与识别算法

为有效并准确诊断出斜拉桥损伤,对基于模式识别的斜拉桥损伤诊断方法进行了研究。选取易于测试出的低阶模态频率和部分关键点竖向振型数据为动力指纹,无需模态扩展或模型缩聚。研究并采用全因子设计进行动力指纹库的创建,可精确评估设定的损伤因子及其交互作用对损伤识别结果的影响。设计并增加了带随机误差的动力指纹库样本集。编制了基于Matlab的模式识别的多种算法,重点研究了精确度高的多层感知器识别算法及其提高该算法预测准确率的装袋集成算法。最后给出一座单塔双跨双索面斜拉桥的多种识别算法的损伤诊断过程和结果,得到一种可包容测试随机误差的高精确度斜拉桥损伤诊断评估模型。     

Fusion method for infrared and visible light images based on image segmentation and stationary wavelet transform

A novel image fusion method based on image segmentation and stationary wavelet transform (SWT) is proposed to improve the visual effect of fused infrared and visible light images. Infrared image is firstly separated into object and background region utilizing Otsu combined with edge detection. Then a multiresolution decomposition using SWT is made to the background region of the infrared image and the visible light image. Neighborhood spatial frequency and absolute value are adopted as fusion rules in low-frequency and high-frequency coefficients. The background fused image is reconstructed by inverse SWT. The final infrared and visible light fused image is obtained by fusing the background fused image and the object region of infrared image base on weighted fusion rule. The experimental results show that the object information of the infrared image is obviously highlighted and the scene information of the visible light image is well represented. The visual effect of fused image is improved efficiently by utilizing the proposed method. The proposed method works better than the traditional Laplacian Pyramid and wavelet transform fusion algorithms in terms of standard deviation, comentropy and mutual information. Experimental results verify its effectiveness.     

频域滤波的水泥路面图像降噪增强方法

针对在水泥路面裂缝图像识别中刻槽和纵向条带状形成的噪声干扰,提出了一种基于频域滤波的降噪增强方法。首先,分析了这两类噪声的频谱特征,并根据空域图像中刻槽的周期性,推导了频域图像中刻槽谱峰的位置,设计了消除这些谱峰的系列滤波器。根据纵向条带状干扰属于低频成分的特性,设计了抑制低频、增强高频的滤波器,经优化滤波器参数达到消除这类干扰的效果。实验结果表明,该方法能够消除两类噪声干扰,同时,增强了裂缝的对比度,为水泥路面裂缝识别提供了理论基础。     

Study on diagnostic method for defects in an IGBT module based on DTW deviations of time series

Insulated gate bipolar transistor(IGBT) is one of the key reliability sensitive components of power electronic equipment. Developing prognostic method for defects in an IGBT module is an important measure to enhance the reliability lever of IGBTs. Therefore,a novel prognostic method based on dynamic time warping(DTW) deviations of time series is presented. This method utilizes variations of DTW of gate voltage series caused by defects to estimate if any defect existes in an IGBT module before breakdown. And application value is verified by test results.     

基于多重表型分析的准确评价高粱抗旱性方法的建立

干旱是制约农业生产的重要非生物胁迫之一,因此作物抗旱性研究具有重要理论和应用价值。本研究对73份高粱材料萌发期和苗期的抗旱性进行了初筛,从中选择15份抗旱材料和6份敏感材料,通过室内和田间试验,获得干旱胁迫下幼苗的株高、叶长、叶宽等形态特征以及超氧化物歧化酶、丙二醛、过氧化物酶等生理生化指标的响应数据,并对高粱的抗旱性做进一步的验证,最终确定抗旱型材料1份,中等抗旱型材料1份,干旱敏感型材料2份。本研究建立了一套基于多重表型分析的高粱抗旱性研究新方法,即综合运用多种数据分析方法(包括隶属函数法、主成分分析法、聚类分析法),结合不同生长环境(田间试验和室内试验)和不同生育期(萌发期和苗期)高粱响应干旱胁迫的多种表型数据(形态特征、生理生化指标),通过多重比较高粱抗旱性评价结果的一致性,旨在系统、高效、准确地判定高粱的抗旱性,为高粱抗旱机制的研究和抗旱新品种的选育提供技术支撑。     

Selection of cereals for weed suppression in organic agriculture: a method based on cultivar sensitivity to weed growth

Cereal cultivars conferring a high degree of crop competitive ability, especially against aggressive weeds, are highly beneficial in organic farming as well as other farming systems that aim to limit the use of herbicides. In this study, thirteen winter wheat cultivars, plus one spring wheat and one winter oat were assessed for their competitive ability at key growth stages, across three seasons. The natural population of weed species was allowed develop without agronomic intervention. Weed suppression ability for each cultivar (S _var) was calculated as the difference between weed growth in plots for each cultivar and the maximal weed growth (W _max) from adjacent uncropped areas. The sensitivity of S _var in response to changes in weed growth (S _var^W) was derived from the linear regression coefficient of S _var plotted against W _max. There was significant variation in S _var between cultivars and strong evidence for cultivars to vary in S _var^W. Amongst groups of cultivars with similar levels of S _var some could be defined as being of higher or lower sensitivity to changes in weed growth. Some cultivars also had relatively good S _var at high levels of weed growth. The use of both weed suppression ability and sensitivity across different levels of weed growth or weed populations has considerable potential for selecting new cultivars suitable for organic agriculture. Ideally new cultivars will be selected on the basis of high S _var and/or low S _var^W. This analysis provides the means to measure sensitivity of cultivar performance across a range of favourable and unfavourable conditions.     

A new method for evaluation of the resistance to rice kernel cracking based on moisture absorption in brown rice under controlled conditions

We developed and evaluated the effectiveness of a new method to detect differences among rice cultivars in their resistance to kernel cracking. The method induces kernel cracking under laboratory controlled condition by moisture absorption to brown rice. The optimal moisture absorption conditions were determined using two japonica cultivars, ‘Nipponbare’ as a cracking-resistant cultivar and ‘Yamahikari’ as a cracking-susceptible cultivar: 12% initial moisture content of the brown rice, a temperature of 25°C, a duration of 5 h, and only a single absorption treatment. We then evaluated the effectiveness of these conditions using 12 japonica cultivars. The proportion of cracked kernels was significantly correlated with the mean 10-day maximum temperature after heading. In addition, the correlation between the proportions of cracked kernels in the 2 years of the study was higher than that for values obtained using the traditional late harvest method. The new moisture absorption method could stably evaluate the resistance to kernel cracking, and will help breeders to develop future cultivars with less cracking of the kernels.