A model for logarithmic image processing

Up to now, image processing and image analysis techniques have borrowed their basic tools from functional analysis: Fourier filtering, differential and integral calculus, and so on. These tools, however, only realize their efficiency when they are put into a well-defined algebraic frame, most of the time of a vectorial nature. Unfortunately, the class of functions modelling ‘images’, commonly referred to as ‘grey tone functions’ does not necessarily present this very type of structure. We present here an operation for the ‘addition’ of two images, with a physical justification in the context of transmitted light. Such an addition permits the construction of the family of ‘positive homothetics' of the grey tone function at hand. The vectorial context sought is well defined: The class of images associated with the class of their grey tone functions naturally becomes the positive cone of an ordered real vector space. Furthermore, the proposed model holds for logarithmic imaging and is compatible with what is known about the human visual process. This model has been called ‘LIP’ (logarithmic image processing model).     

Automatic discrimination of features in grey-scale images

Automatic setting of discriminator thresholds to select features from grey-scale images is desirable to reduce variations due to operator variability, and to adapt to changing illumination or sample characteristics. This can be carried out in several ways. One locates minima in the original brightness histogram. For images in which minima are absent or poorly defined other techniques are preferred. Settings that minimize the change in total feature area or perimeter with changes in the threshold setting are suitable for a wide variety of images. After processing to extract edges a fixed percentage of the number of pixels in the image may be useful.     

面向ICT切片图像复杂轮廓数据的三角网格拼接

针对从ICT(Industrial Computerized Tomography)切片图像中提取出的复杂轮廓数据的三维重建中数据量过多、算法复杂等问题,提出一种简单快速有效的三角网格拼接方法。即采用Freeman链码对断层图像的轮廓数据进行存储和精简,再根据相邻层间轮廓的相似性采用凸包理论来确定轮廓对应点,最后根据轮廓对应点情况对原有的轮廓点集合进行增补后采用最短对角法进行网格拼接。试验结果表明,该方法在保证重建精度的情况下有效减少数据量和计算消耗。     

基于机器视觉的螺纹参数检测

依靠机器视觉技术,对螺纹检测的图形处理技术进行了研究,提出了螺纹多参数的检测方法.介绍了图像处理技术中的预处理、边缘轮廓提取、参数标定等技术,并研制了相应的检测软件.实验结果表明该软件能对螺纹进行实时、高速地检测.     

基于预分割轮廓的CT图像亚体素表面检测方法

针对基于CT(computed tomography)图像检测分析中的点云提取精度与完整性问题,提出一种基于预分割轮廓的高精度、高完整性的亚体素表面检测方法。首先采用Otsu分割算法提取CT图像的体素级轮廓点集,并以此作为粗定位轮廓自适应地生成用于亚体素表面检测的完备感兴趣区域(region of interest,ROI);然后提出一种基于梯度非极大值抑制的表面体素判定方法,避免了梯度阈值选择难题;最后基于3D Facet模型定位亚体素级表面点位置。实验结果表明,该方法能有效改善传统亚体素检测方法的轮廓丢失、伪边严重等问题,轮廓定位误差小于0.2个体素,同时能够取得3倍以上的计算加速比。     

Scanning image detection (SID) system for conventional transmission electron microscope (CTEM) images

A new image detection system has been developed to display transmission electron microscope (TEM) images on a CRT without a video camera system. Deflection coils placed in both the upper space of an objective lens and in the lower space of the first intermediate lens scan a small electron probe simultaneously. The electrical signal acquired through an improved scintillator and a photomultiplier is synchronized with the scanning signal and displayed in a similar fashion to a conventional scanning TEM (STEM) instrument. A preliminary system using a 100 kV conventional TEM (CTEM) equipped with a hairpin-type electron gun, produced an image with a spatial resolution of 1 nm.     

由冠脉光学相干层析图像自动提取血管壁内轮廓

研究了冠脉光学相干层析(OCT)图像血管壁内轮廓的自动提取方法以提高算法的鲁棒性和计算效率.首先,利用冠脉OCT体数据中相邻图像之间的相关性去除鞘管信号对边缘提取的影响.接着,采用射线发射法估计血管中心,并以血管中心作为极点对去除鞘管信号的图像进行极坐标变换.然后,采用分块阈值滤波去除变换后图像的噪声并提取上边缘.最后,做逆极坐标变换得到血管壁的内轮廓.对400多幅测试图像的实验结果表明,该方法可以正确提取冠脉OCT图像的血管壁内轮廓,每幅图像的平均处理时间约为1.2s.该方法不仅能够正确处理血管分叉或鞘管干扰大的图像,而且计算效率高,因此在鲁棒性和处理速度上具有优势.     

基于机器视觉的机械零件图像尺寸自动标注方法

机械零件图像二维几何尺寸自动标注是机械零件视觉检测系统的重要组成部分.对机械零件图像二维几何尺寸自动标注进行研究,提出一种在MATLAB环境下基于机器视觉的直线段、圆和圆弧尺寸自动标注方法.通过分析MATLAB中的坐标关系,进行坐标转换,再根据机械零件图像的二维几何特征,推导出对应的尺寸标注公式,并给出该方法的具体实施步骤,最后通过实验验证了该方法的可行性和有效性.     

应用自调整蛇模型实现提取布氏硬度压痕圆

针对零部件的布氏硬度压痕图像,描述了一种应用蛇模型并结合形状相似性度量从而进行图像中的压痕圆提取的新算法.该新算法使用了改进后的蛇模型,可以利用预先知道的元件压痕的形状信息,基于形状相似性的度量,进行蛇模型的参数自调整,从而对元件的布氏硬度压痕图像进行有效的边缘检测,最终采用Hough变换成功实现了对布氏硬度压痕图像中的压痕圆的识别.实验结果表明,该方法可以实现了蛇模型参数的自适应设置,是一种有效的压痕圆提取方法.     

OpenCV在条烟视觉检测系统中的应用

条烟视觉检测系统是安装在条包生产线上,在条包装箱之前,对其外观缺陷进行检测的一种专用设备。系统可以对生产线上的条包进行实时检测,并及时剔除外观有缺陷的产品,具有高速、高效、准确的优点。该检测系统是集机械、光电、计算机、通信等技术为一体自动化检测系统,系统具有高速、高效、准确的优点,它的使用不仅可以提高烟草生产企业的自动化程度,还可以降低工人的劳动程度,改善工作环境,保证产品的质量。对常见条烟检测项进行了算法设计,并利用某公司的开源计算机视觉库OpenCV进行算法实现,提高了检测速度和精度。     

利用局部方向微分向量一致性的角点检测

在边缘轮廓提取的基础上,提出了一种利用局部方向微分向量一致性的角点检测算法以消除边缘噪声对角点检测产生的不利影响。该算法提取图像的边缘轮廓来降低算法计算量;利用各向异性高斯方向导数(ANDD)滤波器提取每个像素处的方向微分向量并进行幂次变换,以增强向量的各向异性;进而利用相邻像素的方向微分向量构建一致性测度。最后,对同一轮廓上的一致性测度进行均值归一化,得到最终角点测度。实验显示,提出算法的平均角点定位误差为1.52pixel,与对比算法接近;检测准确率分别比点到弦距离累积(CPDA)法、相对局部曲率(HeYung)法提高了58%和5.5%,与归一化残余面积(RA)算法相等,同时角点错检率比HeYung和RA少25.5%和21.6%。提出的算法能准确地检测出真实角点,并具有更小的错误检测率,更高的角点重复率,而且对边缘噪声十分鲁棒。     

基于张量低秩分解和稀疏表示的红外微小气体泄漏检测

为了检测石化工业生产过程中微小气体的泄漏,提出了一种应用红外成像技术的单帧红外小目标检测方法。研究了低秩稀疏分解理论和稀疏表示理论,并提出了一种新的基于张量低秩分解和稀疏表示的小目标检测方法。该方法基于张量分解的形式充分发掘背景矩阵所包含的信息;利用先验知识构造微小气体泄漏的目标字典;同时利用背景的低秩约束和小目标的稀疏表示约束分解出微小气体的泄漏目标。最后基于非精确增广拉格朗日乘子法(IALM),对本文算法进行最优化求解,并通过实验分析比较了本文方法和已有方法的优缺点。结果表明:本文方法的检测效果优于其他已有方法,并且具有较好的ROC(受试者工作特征)曲线,可以满足工业生产中对微小气体泄漏检测的要求。     

Algorithm of detection of moving small-scale objects in a sequence of images

An algorithm is proposed for detection, in a sequence of images, of small-scale objects moving with a known velocity vector, the initial positions of the objects being unknown. Random noise is suppressed by four-channel filtration based on estimating subpixel coordinates of the objects in the sequence of images. Results of numerical experiments are presented as a dependence of the detection probability on the false alarm level for different object sizes and input noise levels.     

结合形态学属性断面与支持向量机的合成孔径雷达图像变化检测

针对传统合成孔径雷达(SAR)图像变化检测方法存在误差大、检测率低等问题,提出了一种基于形态学属性断面(MAP)的SAR图像变化检测方法。该方法利用MAP算法提取差异图像的几何结构特征,构造深入描述图像结构化信息的特征向量空间;在利用阈值法对图像进行分割的基础上,引入偏移因子,实现训练样本的自动选取;最后,用支持向量机(SVM)在多维特征空间中对图像进行变化类与非变化类的分类。实验结果显示:本文算法的检测结果优于基于高斯模型的KI阈值法(GM_KI)、基于广义高斯模型的KI阈值法(GGM_KI)和大津法(Otsu)等3种阈值法的检测结果,Kappa系数保持在0.87以上;当峰值信噪比(PSNR)介于[29,44]dB时,抗噪性能指标保持在0.97以上。这些结果证明了文中方法的有效性和优越性。     

ICT图像实体重建技术中的边缘检测和特征识别

在ICT图像实体重建技术中，需要进行实体轮廓的边缘检测和特征识别。本文采用Sobel算子对图像进行边缘检测，得到图像的边缘点，通过多次过滤进行边缘细化；然后根据轮廓跟踪准则对边缘点进行排序，将图像数据转化为具有单像素宽的轮廓线；最后对轮廓线加以识别，用线段来表示实体的同一特征体素。     

智能三坐标测量机中零件位置自动识别系统

提出了一种基于机器视觉和CAD三维模型的零件自动定位方案，利用虚拟图像匹配的方法确定零件在工作台上的位置和方向。在CAD三维模型中，针对零件的可能放置方式通过透视成像的方法形成多幅虚拟图像。CCD产生的实际图像与各幅虚拟图像进行匹配。确定零件坐标系与机器坐标系的关系，达到定位目的。最后在三坐标测量机平台上实现了该方法，并给出了定位定向实验误差分析。     

High-resolution cell outline segmentation and tracking from phase-contrast microscopy images

Accurate extraction of cell outlines from microscopy images is essential for analysing the dynamics of migrating cells. Phase-contrast microscopy is one of the most common and convenient imaging modalities for observing cell motility because it does not require exogenous labelling and uses only moderate light levels with generally negligible phototoxicity effects. Automatic extraction and tracking of high-resolution cell outlines from phase-contrast images, however, is difficult due to complex and non-uniform edge intensity. We present a novel image-processing method based on refined level-set segmentation for accurate extraction of cell outlines from high-resolution phase-contrast images. The algorithm is validated on synthetic images of defined noise levels and applied to real image sequences of polarizing and persistently migrating keratocyte cells. We demonstrate that the algorithm is able to reliably reveal fine features in the cell edge dynamics.     

A new edge detection method for automatic visual inspection

This paper presents a new edge-detection method which, based on simple arithmetic and logical operations, consists of three procedures: image binarisation, image contraction, and image subtraction. Experiments showed that the proposed algorithm could generate a path one pixel wide with continuous edges, and the proposed algorithm had a better edge-detection accuracy than the 4-connected, 8-connected, and the Sobel techniques. Therefore, the proposed edge-detection algorithm is feasible for use in automatic visual inspection systems.     

基于机械视觉的工件在线实时检测及分拣系统研究

在基于机械视觉及图像处理技术的理论基础上,开发出一套针对机械工件实时在线检测及分拣的系统设备。该系统能够检测出工件生成的二维几何特征轮廓图尺寸且能在软件界面上自动标注,然后分拣机构根据检测出来的尺寸特点进行分拣。介绍了系统的硬件构成及软件系统的实现方法,并对系统的硬件、软件进行设计,最后以钥匙为检测目标检验该系统实际使用的可行性,实践表明该系统能够满足实际生产的使用需求。