关节臂激光扫描系统光条中心提取方法研究

在基于关节臂的线激光扫描系统中,光条中心的提取精度和所用时间直接影响到整个系统的测量精度和实时响应能力。本文在国内外现有算法基础上整合改进,利用迭代阈值法、改进 Sobel 算子和光条修补算法完成光条粗提取,取中心点附近10 pixels利用灰度重心法计算光条中心精确位置。实验证明,该算法在350 mm×350 mm测量范围内,系统测量误差优于40μm,一次提取时间28 ms,确保系统35 f/s扫描频率,满足实时在线测量要求。     

十字线结构光光条提取方法的研究

在结构光视觉检测系统中,光条中心的提取精度直接影响到整个测量系统的测量精度.文章根据红色激光光条的RGB值与背景的差异,提取激光光条,结合像素连通性,采用最小外接矩形法将光条分割,在每个距形区域内采用高斯曲线拟合法提取光条中心,利用最小二乘法拟合光条直线.实验表明该方法具有数据存储量小、实现速度快的特点,且具有一定的抗噪声和断线修补能力.     

一种快速亚像素边缘检测方法研究

提出了一种新的亚像素边缘检测方法。此种方法先经过传统模板算子确定边缘的大致位置，然后用曲线拟合方法求出边缘的精确位置，经过粗、精两次定位，在不牺牲速度的情况下，就可获得亚像素级的定位精度。由于B－样条函数所具有的优越性能，该方法采用它作拟合函数。     

利用计算机改良提取结构光光条中心线的方法

介绍激光三角法常用的结构光光条中心提取方法,分析各种方法的测量原理,介绍这几种方法的缺陷进行说明,提出改进方法——最小二乘曲线拟合,从而提高整个激光三角法的精度。     

一种用于零件定位系统的亚像素边缘检测方法

本文简要介绍了基于计算机视觉的长条形杆件测量系统方案，并说明在该系统中零件图像边缘提取对测量精度起着至关重要的作用。本文结合工程实际在传统亚像素边缘检测方法基础上，提出垂直占空比亚像素边缘提取方法。该算法具有算法简单、效率高、边缘提取准确的特点，已成功用于长条形杆件测量系统中。     

基于高斯积分曲线拟合的亚像素边缘提取算法

针对传统边缘提取算法定位精度低、对噪声敏感等缺点,提出一种基于高斯积分曲线拟合的亚像素边缘提取算法。通过曲面插值求取像素级边缘法截线上各离散点的灰度值,再进行高斯积分曲线拟合,寻找高斯积分曲线的均值点坐标,实现亚像素边缘的精定位。用量块直线边缘进行实验,并与现有亚像素边缘提取算法比较,实验证明基于高斯积分曲线拟合的亚像素边缘提取算法定位精度较高,可以达到1 μm,且算法可靠性高、计算速度快,能够用于高精度测量。     

基于正反图案的亚像素条纹边缘检测方法

为了精确确定格雷码条纹边缘,提出了一种基于正反格雷码图案的直线拟合的条纹边缘定位方法．该方法向测量空间投射条纹宽度相等的正、反色格雷码图案,用摄像机拍摄得到经编码图案调制的强度图像,提取两幅强度图像的条纹边缘区域并对其进行直线拟合,求取两条拟合直线的交点,并将其作为边缘点．采用行扫描的方法求取图像中每一行的边缘点,即可得到图像的条纹边缘．文中具体介绍了测量系统组成、测量及定位原理,分析了具体实现过程．实验结果表明,该方法的相对定位误差小于1％．     

一种光斑图像中心的亚像素提取方法

针对空间发光点所成光斑图像中心的高精度提取问题,本文提出一种亚像素提取方法.该方法建立了具有高斯能量分布的空间发光点的透视投影光斑图像灰度分布数学模型,证明了空间发光点能量中心的透视投影不变性.通过对光斑图像二阶方向导数和自相关函数的分析,获得Hessian矩阵并推导出像素级光斑图像中心的判定条件,再将图像灰度分布函数在光斑图像像素级中心的亚像素邻域内泰勒展开,并求取光斑图像灰度分布曲面的局部极大值点作为亚像素级光斑图像的中心点.与现有经典方法的对比仿真实验结果表明,该算法具有更高精度和更好的鲁棒生,可广泛应用于视觉检测的摄像机标定、三维形貌测量等任务中(类)光斑中心的高精度提取.     

利用二次曲线拟合的CCD图像亚像素提取算法

在图像测量系统中，测量系统的精度与边缘提取的精度成正比关系，普通算法的精度为像素级，现在最常用的亚像素算法是重心法。本文提出了一种利用二次曲线拟合方法的亚像素边缘提取算法，介绍了算法的推导过程，并给出了实验结果。     

用于内表面测量的圆结构光视觉图像处理方法

简要介绍了用于内表面测量的圆结构光视觉系统,分析了内表面圆结构光视觉图像特征,并给出了干扰光条的产生原因。在此基础上,提出了干扰光条的剔除方法与有用光条图像中心的子像素级提取方法。介绍了基于阶跃边缘轮廓模型的边缘提取方法,该方法采用非线性优化方法进行模型参数估计,具有速度快、鲁棒性强、精度高的特点。     

一种高密度光栅结构光编码方法

针对光栅结构光传感器三维测量过程中光条识别问题,提出一种新的金字塔子光栅投射法结构光编码方法。该方法把高密度光栅逐级细分为若干个可容易识别的子光栅,用空间二分编码方法分别对每个子光栅进行编码,并详细介绍了确定光条点码值的方法。把该编码方法应用于光栅结构光传感器中,实现了维纳斯雕像的三维测量,并给出了三维重建结果。结果表明,该编码方法误码率仅为0.75%,特别适用于物体三维轮廓致密测量。     

Enhancement of light extraction efficiency of ultraviolet light emitting diodes by patterning of SiO2 nanosphere arrays

Here we introduce a simple and robust method to improve the light extraction efficiency of ultraviolet light emitting diodes (UV LEDs). Although many previous efforts have focused on etching the GaN surfaces, we employed a simple solution process to texture the GaN surface. Arrays of SiO₂ nanosphere monolayers were spun cast onto a polymer layer, consisting of benzocyclobutene (BCB) resins; subsequently, the bottom half of the SiO₂ nanospheres sunk into the BCB layer. The resulting array formed in a hexagonal-like pattern of ‘nano-lenses’ and the photoluminescence measurement exhibited that these patterns enhanced the light extracting efficiency of UV LEDs by 23%.     

Enhancement of the light extraction efficiency in organic light emitting diodes utilizing a porous alumina film

The light extraction efficiencies of organic light emitting diodes (OLEDs) utilizing various kinds of porous alumina films with different pore diameters were investigated. The OLEDs with the porous alumina film deposited on the glass surface were fabricated to improve their light extraction efficiency. The porous alumina film was fabricated by using a two step anodizing electrochemical procedure. The current densities as functions of the applied voltage do not significantly change, regardless of the existence and the magnitude of the pore diameter in the porous alumina film. The luminance efficiency of the OLEDs increased with increasing pore diameter. The luminance efficiency of the OLEDs utilizing the porous alumina film with a pore diameter of 70 nm was enhanced approximately 9% in comparison with that of the OLEDs without the porous alumina film. These results indicate that highly efficient OLEDs can be fabricated using a porous alumina film with an optimum pore diameter.     

Ultrafast method of fullerenes extraction from carbon condensate

The Soxhlet extraction method (SE) only used in the extraction of fullerenes from carbon condensate (CC) obtained in the fullerene production facility, which in the literature is often referred to as carbon black, does not provide complete extraction. Fullerenes can also be extracted from the CC suspension and the solvent by the mechanical extraction method (ME) and show higher degree of extraction efficiency. This method combined with filtration process takes only a few minutes and allows a significant reduction in solvent consumption. Extraction via the ME method application makes it possible to carry out extraction of fullerenes twenty times faster than by the SE method. It can be used as an independent method, bypassing the extraction by the SE method, and also as an additional one. Application of the ME method allows additional extraction of fullerenes weight percentage from the CC, after extracting fullerenes from it applying the SE method. In this instance the higher fullerenes prevail in the essences extracted by the ME method with their content reaching as high as 39.50?wt%. The obtained samples of fullerene mixtures (FM) are characterized by the methods of high-performance liquid chromatography and mass spectrometry. Analysis of the results shows that the composition of the FM, isolated by the ME method, differs from the compositions extracted via the SE method.     

基于随机子空间法的海洋平台模态特征实时提取方法研究

海洋平台结构模态特征实时提取是其结构健康监测的一项重要手段.针对传统随机子空间算法在有限测点下无法反映结构整体特征的问题,提出了一种改进的协方差驱动随机子空间模态实时提取方法.该方法基于一种新的Hankel元素重构方式,构建了结构实际空间信息与算法的相关性,并有效降低了计算矩阵的维度,从而显著提高了平台模态识别的准确度...     

一种自动提取IVUS图像血管包络的新方法

提取血管内超声(IVUS)图像的血管包络对冠状动脉疾病的诊断有一定的积极意义。本文综合考虑IVUS图像的灰度特征、序列时间特性、先验知识等三类信息,提出一种自动提取血管包络的方法。先由序列时间特性和先验知识减少噪声和伪像干扰,提取出第一帧图像的初始包络;然后用结合梯度、灰度方差、灰度均值信息的B样条GVFsnake对初始包络进行变形得到第一帧的最终包络;最后利用序列图像的时间特性提取后续帧的包络。通过实验表明:综合三类信息的包络自动提取方法在精度和鲁棒性等方面优于以往的方法。     

A method of determining the spatial position of the phase center of log-periodic antennas

A method of estimating the position of the phase center of log-periodic antennas from the phase radiation pattern in the intermediate zone is proposed. __________ Translated from Metrologiya, No. 4, pp. 39–43, April, 2008.     

一种计算大功率LED光源模块器件结温的方法

研究了大功率LED光源模块各器件之间热传导的相互影响,建立了一种基于热阻矩阵的LED光源模块器件结温计算公式,然后以一个由5只1W大功率器件组成的光源模块为例,演示如何通过测量器件的正向工作电压计算热阻矩阵,进而计算各器件的结温,并与现有方法计算值以及红外测温仪实际测量值进行了比较,结果表明：本方法比现有计算方法更为准确,可以用来预测LED光源的使用寿命和系统可靠性。     

Synthesis of mesoporous ZSM-5 from rice husk ash with ultrasound assisted alkali-treatment method used in catalytic cracking of light naphtha

A series of mesoporous ZSM-5 zeolite was synthesized with ultrasonic assisted alkali-treatment technique and their catalytic activity was investigated in catalytic cracking of light naphtha. ZSM-5 zeolite was synthesized from rice husk ash without using any organic template. Effect of alkali-treatment conditions on physicochemical properties of synthesized zeolite was investigated with XRD, FESEM, TEM, N2 adsorption-desorption isotherm, NH3-TPD and TGA technique. It was found that ultrasound energy facilitates the creation of hierarchical structure of ZSM-5 during alkali-treatment. According to XRD analysis, zeolite structure was preserved after 20 min ultrasonic assisted alkaline treatment. However, prolonged dessilication time led to the destruction of MFI zeolite structure. The synthesized ZSM-5 represented highly ordered hexagonal-shape morphology. With increasing alkali-treatment time, the plough land roughness appeared on the surface of zeolite. Comparison of the textural properties samples revealed that the mesopore surface area of alkali-treated samples increased considerably with the increase of ultrasonic assisted alkali-treatment time. Results from catalytic activity tests showed that ultrasound energy has great influence on the activity of ZSM-5. The sample had the highest activity after alkali-treated for 20 min in presence of ultrasound energy which was due to their appropriate hierarchical structure.