基于拉格朗日与牛顿算法的无创变形测量

在传统网格法变形测量基础上,提出基于图像匹配技术的无创变形测量方法.对传统的图像匹配算法进行改进,并采用基于拉格朗日插值与牛顿法的算法进行亚像素位移测量,综合应用虚拟网格的思想与数字图像匹配技术来实现位移与应变的测量,并与网格法进行比较.实验结果表明,采用该算法明显地提高了测量精度与速度.     

关于图像反走样的理论及应用

图像反走样是用于提升图像质量的一个重要环节。导致图像失真的主要原因是由于采样频率过低及重建过滤器的质量低下。在实时渲染的应用中,随着硬件的发展及渲染架构的改变,主流反走样技术逐渐从硬件加速模式发展到后处理模式。纯后处理算法由于缺失子像素信息,无法生成高质量的反走样图像。针对这一问题,混合反走样在保留了后处理反走样效率的同时,通过结合多重采样能提供更精确的边缘处理,并保证图像在时序上的稳定性。     

Lateral performance evaluation of laser micromachining by high precision optical metrology and image analysis

Today several techniques are available for micro-manufacturing. Yet, it is difficult to assess the precision and lateral X,Y accuracy of these techniques. The available accuracy information is usually based on specifications given by machine suppliers. This information is based on in-house laboratory tests performed by dedicated machine operators and within an adapted environment. In practice, the accuracy is likely to vary due to environmental conditions, materials and operator skills. In order to check the specifications in realistic environments the EUMINAfab infrastructure consortium initiated a set of independent high precision onsite verification tests on different laser micromachining installations. In addition to providing performance verification, it gave the participating partners real capability information of their equipment and possibilities to improve machining performance to a higher level. In this study a comprehensive verification test was designed and carried out by using a high precision metrology method for 2D measurements based on subpixel resolution image analysis. This methodology improved our knowledge of the capabilities of three laser micromachining installations, and showed that specifications at single micron levels are hard to obtain.     

基于亚像素边缘检测的二维条码识别

提出了一种基于亚像素边缘检测的二维条码识别算法、能有效地解决边缘模糊对条码识别的影响。以PDF417条码为例研究了基于亚像素边缘检测的二维条码识别算法。首先定位条码位置并在条码中分割出单个码字符号图像。然后根据分割出来的单个码字符号图像讨论r基于亚像素边缘检测的识别算法。实验结果表明基于亚像素边缘检测的识别算法具有良好的性能，显著地提高了条码的识别率，满足了实际使用的要求。     

Estimating subpixel surface temperatures and energy fluxes from the vegetation index-radiometric temperature relationship

Routine (i.e., daily to weekly) monitoring of surface energy fluxes, particularly evapotranspiration (ET), using satellite observations of radiometric surface temperature has not been feasible at high pixel resolution (i.e., ∼10¹-10² m) because of the low frequency in satellite coverage over the region of interest (i.e., approximately every 2 weeks). Cloud cover further reduces the number of useable observations of surface conditions resulting in high-resolution satellite imagery of a region typically being available once a month, which is not very useful for routine ET monitoring. Radiometric surface temperature observations at ∼1- to 5-km pixel resolution are available multiple times per day from several weather satellites. However, this spatial resolution is too coarse for estimating ET from individual agricultural fields or for defining variations in ET due to land cover changes. Satellite data in the visible and near-infrared wavelengths, used for computing vegetation indices, are available at resolutions an order of magnitude smaller than in the thermal-infrared, and hence provide higher resolution information on vegetation cover conditions. A number of studies have exploited the relationship between vegetation indices and radiometric surface temperature for estimating model parameters used in computing spatially distributed fluxes and available moisture. In this paper, the vegetation index-radiometric surface temperature relationship is utilized in a disaggregation procedure for estimating subpixel variation in surface temperature with aircraft imagery collected over the US Southern Great Plains. The disaggregated surface temperatures estimated by this procedure are compared to actual observations at this subpixel resolution. In addition, a remote sensing-based energy balance model is used to compare output using actual versus estimated surface temperatures over a range of pixel resolutions. From these comparisons, the utility of the surface temperature disaggregation technique appears to be most useful for estimating subpixel surface temperatures at resolutions corresponding to length scales defining agricultural field boundaries across the landscape.     

摄影测量中一种新的编码点识别和提取算法

描述一种符合摄影测量要求的编码点,并提出一种新的识别和提取编码点算法。这种编码点有15位编码,编码的数量达到2191,完全能够满足摄影测量的要求。通过CannY算子边缘提取和椭圆拟合,极大地提高了识别的正确率,能够实现解码的高鲁棒性。编码点中心坐标的精度达到亚像素级。该解码方法简单,且编码点识别率极高。     

利用数字图像处理技术测量直齿圆柱齿轮几何尺寸

研究了采用数字图像处理技术实现对直齿圆柱齿轮几何尺寸的非接触式测量.首先对CCD摄像机拍摄的图像进行畸变校正与滤波去噪,然后对图像进行阈值分割,采用基于数学形态法的四邻域腐蚀来获得单像素宽的图像边缘.针对齿轮的边缘轮廓形状特征,运用重心法、最小二乘拟合、Bresenham画圆法和Hough变换等原理,建立了测量齿轮齿数、模数、中心孔半径、齿顶圆半径、齿根圆半径和变位系数等齿轮几何尺寸参数的测量算法.使用1280×1024的CCD摄像机及黑白采集卡对分度圆直径为50mm的直齿圆柱齿轮进行测量实验,与人工测量值对比,绝对误差小于一个像素,在各测量参数中,最大尺寸的齿顶圆半径绝对误差值为13μm.研究结果表明:因为CCD摄像设备的分辨率越高、被测物体的尺寸越小,则测量精度越高,且为线性关系,所以,使用高分辨率CCD摄像设备并配备光学放大系统,可实现对高精度和微小直齿圆柱齿轮几何尺寸参数的测量.如使用1280×1024像素以上的CCD摄像设备测量分度圆直径5mm以下的直齿圆柱齿轮时,测量精度可在2μm以下.     

高精度贴片机自动定位的快速算法

在简要介绍高精度自动倒装贴片机的原理和实现过程的基础上,阐述了一种快速的高精度的自动定位算法。该算法的基本思想是先用LOG算子得到像素级边缘,再用插值方法得到其亚像素边缘,然后用SUSAN角点检测法提取精确的特征点,最后用VanWamelen算法实现快速点模式匹配。它能在输入标记缺损或遮挡的情况下快速有效完成对准,而且精度较高。     

A timely detection of a coated board streak defect in subsampling conditions using monochrome vision system

This paper presents a study of an important improvement in an already installed monochrome vision system for detection of defects during coated board manufacture. This improvement allows the inspection system to detect targeted defect in subsampling conditions (i.e. Nyquist theorem is not satisfied). This improvement is accomplished entirely in the image processing software without the need for additional changes in the vision hardware. Processing software is optimized for SIMD instruction set requiring no improvement in the processing hardware. Results obtained by the use of this improved algorithm provide real time detection of the targeted defect that in turns enables elimination of the waist coated board.     

应用机器视觉精确检测工件尺寸的研究

本文在介绍机器视觉检测系统的组成和工件尺寸检测原理的基础上,以高精度尺寸的标准块为例,详细介绍了工件图像滤波,边缘检测和亚像素边缘提取,并对亚像素边缘特性和规律进行了简单的分析.实验检测结果表明,对不超过40mm的工件,尺寸检测误差不超过0.1mm;改进系统测量方法后,检测误差不超过0.01mm.