基于ROSST的列车轴承轨边声学信号校正与诊断研究

由列车道旁麦克风采集的列车轴承故障声学信号存在多普勒畸变现象,需要在信号处理过程中进行校正和相关诊断。利用重分配算子同步压缩变换(Reassignment Operator and Synchro Squeezing Transform,ROSST)获取高分辨率时频分布,通过脊线提取、莫尔斯声学理论和非线性拟合得到信号的多普勒畸变参数,再对原信号重采样,消除多普勒畸变。将其应用于仿真信号和列车轴承多普勒畸变故障实验信号的校正与诊断,验证了该方法的有效性。     

基于自制标尺的C臂X光全景图像拼接方法

在对C臂X光图像和现有的医学图像匹配方法分析的基础上,提出了一种基于自制标尺的C臂X光图像拼接方法.该图像拼接方法分为失真校正、图像匹配和图像合成三步,其中图像匹配是算法的关键.首先采用双线性变换模型进行失真校正;然后基于自制标尺,采用图像灰度匹配法对校正后的C臂X光图像进行图像匹配;最后将其余各幅图像间的匹配矩阵均换算成以第一幅图像为基准,据此合成出全景图像.通过尸腿骨X光图像拼接实验验证了该方法的准确性.     

畸变测量中应用窗口傅里叶变换载频条纹分析

为了测量光学成像系统的径向畸变,提出了基于伸缩窗口傅里叶变换空间载频条纹相位分析的测量新方法.畸变分布测量转化为调制相位测量.首先,将纵向朗奇基准光栅作为模板,通过成像系统成为变形光栅即畸变像.接着采用伸缩窗口傅里叶变换提取畸变载频光栅条纹中心无畸变点的基频和相位信息,获得理想无畸变像的基频成分,然后对变形载频光栅条纹进行频谱分析,滤波提取基频信息、逆傅里叶变换、相位解包,提取径向调制相位分布,计算畸变图像的径向位置畸变分布.最后利用该径向位置畸变分布规律和双线性插值灰度重建对畸变图像进行校正.详细的理论分析和实验结果证明,该方法是可行的.     

基于光线投影的医学影像体绘制加速技术

针对标准光线投射算法计算量大、速度慢的特点,本文提出一种满足医学图像实时需要的体绘制加速新算法.利用图元在矩阵变换前后呈均匀变化的特性,递推得到除6个顶点外的其余重采样点的物体空间坐标,大大减少矩阵运算量;同时利用包围盒技术避免对空体元的采样,通过将Bresenham算法扩展至三维使投射光线的体元化与重采样参数的计算一次完成,极大地加速光线投影的效率.大量的对比实验均表明,本文算法和传统标准算法具有相同图像质量,体绘制的速度提高了2～3倍.     

全景图像实时校正系统的设计与实现

利用FPGA作为可见光全景图像处理器,使用硬件畸变校正系统,以达到对全景图像畸变场景的实时校正.全景图像畸变校正核心问题是像素空间位置校正,以CORDIC算法为基础,设计了全景环形图像展开算法,实现了全景环形图像的展开.设计采用ITU-656标准数字视频格式,用VHDL硬件描述语言实现整个校正算法的FPGA设计,试验结果表明,能实现全景环形图像的360.实时展开.     

分解算法的丢失数据射影重建

在射影重建问题中,广泛应用的分解算法尽管能够平等地看待所有的图像,但它们要求全部空间点在所有的图像上都可视.为了克服这一限制,该文提出了一种基于分解算法的射影重建方法.该方法迭代地估计空间点、投影矩阵、深度因子和丢失的数据.投影矩阵和空间点通过奇异值分解被求取.深度因子则根据每幅所有图像点构成的行向量生成的线性子空间和与射影重建空间点构成的行向量生成的线性子空间是同一线性子空间的特性被估计出来.模拟数据和真实图像的实验结果表明,该射影重建方法具有重投影误差小,收敛性好和实用性强的特点.     

一种显微CT系统图像畸变校正方法

基于透镜耦合的光耦探测器由于其高分辨力、倍率选择多样化的特点,被应用到显微CT系统中.然而,光耦探测器的成像畸变影响了显微CT系统的成像质量.为此提出一种适用于显微CT系统的图像畸变校正方法.建立理想图像和实际畸变图像的多项式函数关系,利用特征点坐标求取畸变系数,并利用畸变系数对图像进行畸变校正.实验结果表明,所提方法能够有效地校正畸变,校正后系统的空间分辨力由原来的8.5μm提升到4.2μm.标准网格板图像校正后网格线的弯曲得到明显改善,网格间距趋于平均,离散程度变小.同时,该畸变校正算法复杂度低,具有较高的实用性.     

三次样条插值和中值滤波技术在PDF417译码过程中的应用

在基于图像译码方式完成PDF417译码的过程中,由于通过摄像头采集的条码图像存在畸变、噪声等现象,在条码辨识前必须对图像进行预处理,使条码图像便于分割处理。针对条码采集过程中图像存在的几何畸变,本文采用了新的几何失真校正方法－三次样条插值校正和中值滤波技术。实验结果证明了该方法在完成条码图像几何畸变的校正上起到了较好效果,从而为正确译码打下基础。     

基于扫描图像几何畸变的算法研究

目的为了改进和优化适用于扫描图像的几何畸变校正算法。方法通过分析扫描仪产生几何畸变的原因,总结现有图像畸变校正方法,并与现有的图像处理软件——Photoshop软件相比较,建立合适的数学模型和算法流程。经由图像预处理,利用直方图均衡化法使图像增强,并用凯尼算子进行边缘检测,然后用雷顿算子提取图像轮廓,最后采用连接点法校正畸变图像,并用Matlab仿真加以实现。结果文中算法处理迅速,不需手动处理,校正后平均相似度可达0.95,效果优良。结论预处理后的扫描图像经点连接法进行几何畸变校正,作为一种可以快速处理扫描几何畸变校正方法的补充,算法适用于扫描图像的几何畸变,具有重要的实用价值和研究意义。     

自动修正拍摄俯仰误差的鱼眼图像校正方法

现有鱼眼图像校正法仅针对单幅图像,无法提供足够信息来自动修正由于相机抖动或环境干扰等因素造成的形变误差.本文利用图像拼接时的特征点,计算出待拼接图像之间几何变换关系,进而准确估计出图像的俯仰角度,基于此提出了一种精确校正鱼眼图像的方法.该方法的主要特点能恢复校正图像的垂直不变性,增强校正图像之间的对齐程度,使得全景拼接中变换矩阵求解变得更为简单,增强融合效果.实验证明,本文校正算法可以消除由于形变误差造成的图像拼接中的重影,增强拼接图像的效果.     

摄影测量中相机倾斜引起的射影失真研究

根据成像原理,建立了理想光学系统成像的数学模型。通过分析该模型中物与像的对应关系,得出了射影失真产生的根源是光学系统放大率的不恒定,并与相机的外参数:倾角及视场的大小有着非线性的变化关系。提出了正射影失真与负射影失真的概念,对其进行了详尽的理论分析,并给出了数学模型及分布规律。依据上述结论,分析了二维射影对应与仿射对应在摄影测量应用中的实质区别,并利用物像平面间的射影对应关系,使试验中的测量精度达到了测试的要求。     

Binary Image Steganalysis Based on Distortion Level Co-Occurrence Matrix

In recent years, binary image steganography has developed so rapidly that the research of binary image steganalysis becomes more important for information security. In most state-of-the-art binary image steganographic schemes, they always find out the flippable pixels to minimize the embedding distortions. For this reason, the stego images generated by the previous schemes maintain visual quality and it is hard for steganalyzer to capture the embedding trace in spacial domain. However, the distortion maps can be calculated for cover and stego images and the difference between them is significant. In this paper, a novel binary image steganalytic scheme is proposed, which is based on distortion level co-occurrence matrix. The proposed scheme first generates the corresponding distortion maps for cover and stego images. Then the co-occurrence matrix is constructed on the distortion level maps to represent the features of cover and stego images. Finally, support vector machine, based on the gaussian kernel, is used to classify the features. Compared with the prior steganalytic methods, experimental results demonstrate that the proposed scheme can effectively detect stego images.     

基于改进传递矩阵法的环肋圆柱壳固有振动分析

为了简化传统传递矩阵法中状态向量一阶微分方程复杂推导和得到不同边界条件下环肋圆柱壳的振动特性,基于Flügge壳体理论,通过采用改进传递矩阵法改进状态向量的选取,直接快速地从振动方程推导出圆柱壳结构场传递矩阵,并对场传递矩阵使用精细积分求解。根据环肋和壳体连接处变形连续条件导出环肋处点传递矩。最后通过自由、简支、固支三种不同边界条件下环肋圆柱壳固有频率计算结果与有限元计算结果进行对比,验证了改进传递矩阵法进行环肋圆柱壳振动分析有效性和适用性。     

Theoretical analysis and reduction of the quantization noise effects in the SA‐DCT

We present a detailed mathematical analysis of two important problems in object‐based video coders that use shape‐adaptive discrete cosine transform (SA‐DCT). One is the annoying mean distortion effect caused by the quantization process and the other is the quantization noise dependence on the object shape (and not only on the quantizer characteristics). A general expression for the cross‐correlation matrix of the image error magnitude is presented, as well as one for the particular case of white quantization noise. In the latter case, four different implementation options are examined. We propose new strategies to minimize (or even eliminate) the two problems considered in our mathematical formulation. The results of our experiments show that the proposed scheme outperforms those obtained with other strategies described in the literature. © 2005 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 14, 238–245, 2004; Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/ima.20030     

起重机主梁腹板变形自动矫正执行器设计

起重机主梁生产线中下盖板与形梁组对时,用人工矫正方法矫正腹板波浪变形会在主梁腹板上留下焊疤,并增加没有附加值的打磨操作。为了改善这一状况,并高效配合机器人实施定位焊,采用磁力吸附原理进行腹板变形的自动矫正。首先对形梁腹板变形进行测量,并拟合出变形曲线,不同格子间的变形均为凹变形,最大变形值为24mm。接着利用有限元方法分析了磁铁布置方式对矫正效果的影响:在同样的磁铁吸附应力和布置方式下,大筋板间的间距越大,矫正效果越明显;而在同样的大筋板间距下,初始凹变形小有利于矫正;对所有的格子间,永磁铁距腹板下边缘越近,矫正效果越明显,同样的磁铁(长宽比大于1),横放比竖放的矫正效果明显。最后研发了自动矫正执行器的物理样机,其由吸附单元、推拉单元、结构单元、传感单元和控制单元组成,并进行了现场试验,测试结果满足矫正工艺的精度要求(±2mm),同时也验证了有限元分析结果的准确性。主梁腹板变形自动矫正执行器的研制实现了腹板变形识别、腹板临时连接的建立、矫正过程力感知和矫正运动可控。该装置实现了机器人定位焊前的腹板变形自动矫正,对前道工序的焊接工艺优化具有一定意义。     

Design of automatic rectification effector for deformation of crane girder web北大核心CSCD

When the bottom plate and the II-beam are assembled in the crane major beam production line, the manual rectification process will produce scars on the girder web that need additional polishing operation. In order to improve the condition and cooperate with the robot to deal with the tack weld, the method of magnetic adsorption was used to realize the automatic rectification for the girder web deformation. Firstly, the deformation of the girder web was measured, and the deformation curves were fitted. All deformations were concave, and the max value was 24 mm. Secondly, the effect of the layout of the magnet was analyzed by the finite element method. Under the same magnetic adsorption stress and the same magnet layout, the larger the span between the big rib plates were, the better the rectification effect became. And under the same span between the big rib plates, the small initial deformation was beneficial to rectify. To all grid space, the nearer the distance from the lower edge of the girder web was, the better the rectification effect became. The horizontal layout was beneficial to rectify as compared with the vertical layout to the same magnet (the ratio of the length and the width was greater than 1). Finally, the physical prototype of automatic rectification effector was developed, which consisted adsorption unit, push-pull unit, structure unit, sensing unit and control unit, and the field testing was carried out. And the results of the testing were satisfied with the accuracy requirement(±2 mm) of the rectification process. Moreover, the results of the testing also verified the conclusion of the finite element analysis. As a conclusion, the development of the automatic rectification effector realizes the identification of the deformation of the girder web and the temporary establishment with the girder web, the force perception in the rectification process and the control to the rectification motion. The device realizes the automatic rectification of the web deformation before the robot positioning welding, which is of great significance to optimization of welding technology of the previous process.     

无地面控制点航空图像无缝拼接算法

针对航拍图像拼接中,因为航带中航片拼接误差积累导致拼接后期图像发生扭曲问题,本文提出一种利用捆绑调整技术削弱航片拼接过程中误差累积。该算法采用SIFT（Scale Invariant Feature Transform）进行特征点提取和匹配;结合改进的RANSAC（Random Sample Consensus）对特征点进行提纯,剔除外点;由过滤后的特征点通过最小二乘法计算图像间的单应性矩阵,在此基础上运用捆绑调整法整体优化单应性矩阵进行图像间的拼接,解决了拼接后期图像扭曲问题。最后,通过动态加权的融合方法实现图像接缝处平滑过渡。为验证该算法的有效性,选用真实无人机航拍序列图像进行拼接实验,取得良好的拼接效果。     

Design of lenses with curved Petzval image surfaces

Abstract

Design strategies have been devoted to simplify and miniaturize optical systems. In this paper, by constraining image surface to coincide with the Petzval surface, we achieve a compact f/2.8 lens system design with a curved Petzval image surface. Arc distortion is proposed to accurately measure the distortion relative to a curved image surface. The optical performance of our curved image surface lens is analysed and compared. Results show that modulation transfer function (MTF) of our curved Petzval design over 69% at 100 cycles/mm for entire fields is achievable with 100-mm effective focal length, 40º full field of view, >92.4% edge relative illumination, <0.5% arc distortion. Comparisons with a traditional lens with a planar image plane demonstrate that a curved Petzval image surface is an excellent strategy to simplify and miniaturize optical systems, compensate field curvature and benefit astigmatism correction, increase off-axis illumination and improve MTF. Furthermore, the lens with a curved Petzval image surface has a more uniform optical power distribution and greater degree of lens symmetry.     

基于SIFT变换和RBF神经网络的彩色全景图拼接算法

目的 为了复原存在平移、色彩差异、旋转、形变等问题的全景图,提出一种结合SIFT（尺度不变特征变换）和RBF神经网络的彩色全景图拼接算法。方法 通过SIFT算法匹配出两子图中对应的特征点,利用仿射变换解决图像间的旋转和形变问题,采用RBF神经网络纠正子图的色彩差异,最后利用权值矩阵融合技术实现重叠区域的融合。结果 文中算法在拼接效果上优于其他算法,其拼接效果DoEM值为0.902,图像重叠区域过度平滑,有效地避免了融合区域的亮度块或亮度线。结论 该算法效果好,可解决全景图复原过程中多方面的难题。