Characterizing cavitation erosion particles by analysis of SEM images

Wear particles produced by vibratory cavitation erosion on stationary-aluminum Al-99.999 were analyzed using scanning electron microscope (SEM), forming a database for further analysis. The particle morphology features were first clarified based on the characteristic stages of the vibratory erosion rate-time pattern. Next, size, area, perimeter and shape factors (elongation and roundness) were determined for eroded particles, using image analysis software. In incubation period, the particles have distinctive characteristics which differed from that for the subsequent periods. These characteristics include the value of longitudinal ratio and roundness factor, limit size range and morphological features such as lamella shape, folding, curving with one of the particle surfaces as the virgin surface. In acceleration, steady-state and attenuation periods, the particles have a wide size range and larger thickness compared with that for the incubation period. The maximum particle size reached about 360 μm in acceleration and steady-state period. For all the cavitation erosion rate periods, the particles were out of sphericity and they have a roundness factor larger than 2. Detailed surface characteristics of the particles produced during cavitation erosion is significant and can open ways for monitoring the cavitation erosion progress.     

Automated recognition of atypical nuclei in breast cancer cytology specimens by iterative image transformations

In order to develop an objective grading system for nuclear atypia in breast cancer, an image analysis technique has been applied for the automated recognition of enlarged and hyperchromatic nuclei in cytology specimens. The image segmentation algorithm, based on the ‘top hat’ image transformation developed in mathematical morphology, is implemented on the LEYTAS automated microscope system. The performance of the segmentation algorithm has been evaluated for fifty malignant and eighty-five benign breast lesions by visual inspection of the displayed ‘flagged’ objects. The mean number of flagged objects per 1600 image fields for breast cancers was 887 (range 0–7920) of which 87% consisted of single, atypical nuclei. For benign lesions the mean number was 30 (range 0–307) of which 20% were single nuclei. By adaptation of the ‘top hat’ parameter values, a more extreme subpopulation of atypical nuclei could be discriminated. The large interspecimen variation in the breast cancer results was related to differences in DNA content distribution and mean nuclear area, determined independently with scanning cytophotometry, and to some extent with the histological type.     

Combined flow cytometry and image cytometry of the same cytological sample

Flow cytometry and image cytometry, two measuring techniques in the field of analytical cytology, can be used sequentially on the same cytological sample. Cells stained with a fluorochrome for the determination of for example, DNA or RNA content are first analysed in suspension by flow cytometry. The results of the fluorescence analysis of the individual cells are presented after data processing as frequency histograms of the DNA or RNA content of all the cells of the sample. In these histograms certain cell populations such as those with an increased DNA content are defined and these are then selected for further investigation. This is achieved by sorting cells of interest into centrifugation buckets by means of electrostatic deflection of the droplets containing such cells. Sorted cell populations are then centrifuged on to glass slides and stained according to the acriflavine Feulgen–SITS staining procedure, a quantitative method for DNA and protein. Image cytometry of these stained cells is performed with a computer controlled television based image analysis system (LEYTAS). With this system abnormal cells with elevated DNA content or increased chromatin contrast are automatically detected, thereby eliminating almost all artefacts and normal cells. Subsequently detected objects are stored in grey-value memories after the automated analysis for visual examination by the cytologist. The possibilities of combined flow cytometry and image cytometry are illustrated in typical examples in the field of cervical, bladder and mammary cytology.     

Powder sampling and characterization by digital image analysis

The knowledge and understanding of powder shape and size are essential to the improvement and control of several industrial processes. This paper describes an offline methodology used in the characterization of powder properties based on digital image analysis. The method provides a reliable sample preparation consisting of a mixing operation and a statistical test validation. A versatile digital image procedure is presented: image calibration, enhancing and clustering permit to automatically detect particles and retrieve 2D properties. Shadow casting is applied to achieve the third dimension. The method can deliver statistically significant size and shape information in a single measurement characterizing the salient parameters of powder. The potential of the proposed method is shown with its application to some industrial powders. As is suggested it fulfils the identified needs of product and process monitoring applications for technological/industrial manufacturing, for quality control and improvement purposes.     

Ultrastructural morphometric image analysis applied to estimation of condensed chromatin in normal and neoplastic lymphocytes

Currently, quantitative studies of malignant lymphoma are being performed in an attempt to improve the classification of non-Hodgkin's lymphoma (NHL) for diagnostic, prognostic, and therapeutic purposes. Morphometric image analysis is one method that can be employed in cases of NHL to obtain objective data of nuclear parameters; condensed chromatin being a compartment of the nucleus best measured at the ultrastructural level. This report assesses similarities or differences in the amount, distribution, and arrangement of condensed chromatin in nuclear profiles of normal and neoplastic lymphocytes in human surgical biopsy specimens. Morphometric data derived from electron micrographs of lymphocytes in germinal centers of lymph nodes with reactive hyperplasia (three cases) and small cell types of NHL two examples of malignant lymphoma, well differentiated lymphocytic type (ML, WDL) and three cases of malignant lymphoma, poorly differentiated lymphocytic type (ML, PDL) are compared. Results indicate that the distribution of condensed chromatin, i.e., the size of aggregates, and their spatial placement within the nucleus varies more than the amount (both mean area per profile or mean volume) of this nuclear parameter, and that this applies to normal as well as neoplastic lymphocytes. When a series of condensed chromatin parameters were statistically compared, no major differences could be detected between lymphocytes in normal tissues and those in ML, WDL and ML, PDL, but considerable differences were found in each of the nuclear morphotypes in the individual cases within the groups. This degree of variation in nuclear characteristics within normal tissues and the two lymphoma categories has not been previously recognized. Clearly, the technique of morphometric analysis, as applied to electron micrographs, can provide new and useful data that must be appreciated if classification schemes currently used in NHL are to improve and reflect biologic considerations.     

基于模式识别的等离子射流形态特征的研究

射流形态是熔射过程中最直观的射流信息,它是熔射各工艺参数稳定与否的直接反映。本文提出基于CCD对氩氮等离子熔射射流进行图像采集,利用数字图像处理方法提取射流形态,以射流狭长度来描述射流形态特征的识别方法。实验研究结果表明:射流狭长度特征可以很好地反映不同工艺条件下射流的形态特点;通过实验分析主要工艺参数对该特征的影响规律,得到输入电压对该特征有主要影响的结果;结合相应工艺条件下粒子飞行行为检测实验结果,对该射流诊断系统进行验证,表明该系统可实际应用于RSTP的工艺分析中。     

Application of confocal scanning laser microscopy for an automated nuclear grading of prostate lesions in three dimensions

Confocal scanning laser microscopy provides the opportunity to obtain three-dimensional (3-D) images by piling up consecutive confocal planes. This technique was applied to capture 3-D images from 100-μm-thick tissue blocks from prostate lesions (hyperplasia, dysplasia, adenocarcinomas). Automated methods were implemented to perform a nuclear grading of 3-D cell nuclei from these specimens. Special attention was focused on the development of a new approach to 3-D chromatin texture analysis. This method uses mathematical morphology operations to tessellate the chromatin into homogeneous domains. The nuclear features (volume, shape, texture) were subjected to a discriminant analysis. Using a set of five features, the classification of cell nuclei yielded an accuracy of 963%. The results indicate the potential of 3-D imaging and analysis techniques for an automated nuclear grading of prostate lesions.     

Characterization of ovarian nuclei with the parameter of power ratio

The nucleoli and chromatin clumps of ovarian cells contain important features in discriminating malignant cells from normal ones. In geometric properties, the ovarian nucleoli and chromatin clumps appear as irregularly shaped dark spots in the nuclear images from specimens immunohistochemically stained with antibody to Mib-1. Malignant cells often have more active and larger nucleoli and chromatin clumps. However, estimating the size of the nucleoli or chromatin clumps is a difficult task since it is not easy to recognize and accurately separate the regions of nucleoli and chromatin clumps from the rest of the nuclei that are highly irregular and variant in contents and intensities. In this paper, we develop a method to derive a parameter called power ratio that is proportionally related to the size of nucleoli and chromatin clumps based on an ideal nuclear model without the region segmentation of nucleoli or chromatin clumps. Results of characterization of the parameter and comparison between malignant and normal cells are provided.     

基于数学形态变换的转子故障特征提取方法

基于非线性数学形态变换提出旋转机械故障特征提取的新方法.由数学形态变换构成的形态滤波器可以有效地提取出信号的边缘轮廓以及形状特征,通过选取不同长度的形态结构元素,采用组合形态滤波器将旋转机械故障信号分解到不同频带上,故障信号被分解成基频成分、故障成分及高频噪声三部分,在分解过程中,信号长度没有减少,没有信息的丢失;将分解得到的故障成分单独提取出来进行分析,可以更准确描述故障特征;对实际碰摩故障信号进行形态学分解后,提取出故障成分,采用Hilbert-Huang变换(Hilbert-Huang transform,HHT)对分解前后的信号进行对比分析,验证了方法的有效性,表明基于形态变换的信号特征提取可以更准确刻画故障的非平稳特性,提高了分析效果,并具有计算简单、快速的优点.     

Reproducibility and accuracy of interactive segmentation procedures for image analysis in cytology

The segmentation of nuclear images is a crucial step in the development of procedures using image analysis for the cytological diagnosis of cancer. The purpose of this study is to evaluate the reproducibility and accuracy of several interactive segmentation methods which can be used in this context. Four methods were studied: a thresholding-based method enabling selection of intensity histogram contrast and brightness, manual tracing with a stylus, and arc- and ellipse-fitting routines. Features of nuclear size and shape were derived from nuclei segmented on repeated occasions by several individuals. Variance component models provided a statistical framework for evaluating the intraobserver and interobserver variability of these measurements in terms of their intraclass correlation coefficients. Of the methods tested, the arc-fitting segmentation method gave the most reproducible results, and thresholding the least. Reproducibility was generally very high both between individuals and for repeated segmentations by a single individual. Accuracies of area measurements for the various methods, as determined with respect to point counting, paralleled the reproducibilities of the methods. Sample size requirements were observed to be more dependent on the biological variability of the tissue sampled than on the particular segmentation method or on the number of individuals performing segmentation.     

The Use Of The Texture Analyser To Study Sinter Structure: Application To The Morphology Of Calcium Ferrites Encountered In Basic Sinters Of Rich Iron Ores

Quantitative image analysis applied to iron ore sinters gives complementary information to chemical analysis such as quantitative mineralogical analysis, and also quantitative geometric properties of crystals and of porous networks (size distribution, repartition). Spatial disposition of sinter components produced during crystallization builds different textures which are well known to mineralogists. The types of textures encountered can be related to initial composition, cooling conditions and to physical properties of the sinter (mechanical strength, reducibility). In order to estimate the relative importance of each texture in a sinter, measurements made using a pattern recognition procedure are required. Algorithms for automatic pattern recognition were established for calcium ferrites texture encountered in rich iron ore basic sinters. They depend on a specific treatment of morphological parameters estimated inside fields of measurement of the texture analyser. Furthermore, mathematical models of multicomponent structures based on mathematical morphology, are used for different calcium ferrites textures. The parameters of these models are estimated from experimental measurements made with a texture analyser. The present study is a part of a research programme made by IRSID (Institute of Research of French Steel Industry) on the structure of iron ore sinters in order to explain variations of their physical properties related to their behaviour in the blast furnace.     

基于最小生成树的显著性自适应形态学结构元素构造方法

经典数学形态学运算采用固定大小和形状的结构元素处理整幅图像,由于图像内容的多样性以及目标结构的复杂性,容易导致处理后的图像形状发生改变且丢失部分信息.为此,提出了一种基于最小生成树(Minimum spanning tree,MST)的显著性自适应形态学结构元素构造方法.首先,计算图像梯度,通过非极大值抑制(Non-maximum suppression,NMS)得到边缘图像,对边缘图像进行倒角距离变换,得到显著性图(Salience map,SM).然后,通过计算SM的极大极小值确定结构元素半径,并在SM上计算MST.最后,利用计算得到的半径构造出一种形状和大小随输入图像局部特征自适应变化的结构元素.利用该自适应结构元素对腐蚀、膨胀、开和闭等基本形态学算子进行了重新定义,并且与经典形态学算子做了仿真对比.结果表明,该方法能够充分利用图像的局部特征,在图像结构保持以及图像滤波等方面都具有较好的处理结果.     

基于最小生成树的显著性自适应形态学结构元素构造方法

经典数学形态学运算采用固定大小和形状的结构元素处理整幅图像,由于图像内容的多样性以及目标结构的复杂性,容易导致处理后的图像形状发生改变且丢失部分信息.为此,提出了一种基于最小生成树(Minimum spanning tree,MST)的显著性自适应形态学结构元素构造方法.首先,计算图像梯度,通过非极大值抑制(Non-maximum suppression,NMS)得到边缘图像,对边缘图像进行倒角距离变换,得到显著性图(Salience map,SM).然后,通过计算SM的极大极小值确定结构元素半径,并在SM上计算MST.最后,利用计算得到的半径构造出一种形状和大小随输入图像局部特征自适应变化的结构元素.利用该自适应结构元素对腐蚀、膨胀、开和闭等基本形态学算子进行了重新定义,并且与经典形态学算子做了仿真对比.结果表明,该方法能够充分利用图像的局部特征,在图像结构保持以及图像滤波等方面都具有较好的处理结果.     

基于改进形态学梯度的齿轮边缘检测

经典的边缘检测算子算法更多地采用Prewitt算子、LOG算子和Canny算子等在空域中进行。数学形态学在图像处理过程中不涉及变换,其原理是利用结构元素检测图像,并直接处理图像的特征信息。在形态学梯度的基础上,提出了一种基于改进形态学梯度的图像边缘检测算法,选取合适的形状以及尺寸的形态学梯度结构元素,并组合使用以检测出较理想的图像边缘信息。试验结果表明,该算法在含噪图像中能较好地保存图像边缘信息,有更好的定位精度和抗噪性。     

基于支持向量回归的小尺寸零件精密测量

为提高小尺寸零件测量精度和速度,提出了基于支持向量回归(SVR)的小尺寸零件精密测量方法。系统以齿形链板为研究对象,对其主要参数进行测量。系统采用透射照明方式,使用A102FCCD数字摄像头采集齿形链板的图像,经过IEEE1394数字接口卡传输到计算机。对含有噪声的原始数字图像实施中值滤波降噪、二值化,轮廓提取及图像旋转等处理,使图像转变成易于检测的单像素宽边缘信息。然后根据齿形链板长度与宽度比例确定待检测区域,以待检测区域内的边缘轮廓上的各像素点构成对应线段的训练集,进行支持向量回归,获得具有亚像素表示的各检测线段的回归函数,并据此对齿形链板的主要参数进行测量。最后,对测量误差进行了分析。测量结果满足零件的公差要求,测量精度可达2 μm。理论分析及实验结果表明,该方法测量速度快,测量精度高,同时对图像平面内旋转、尺度变化、噪声等具有较强的鲁棒性。     

Segmentation of confocal microscope images of cell nuclei in thick tissue sections

Segmentation of intact cell nuclei from three-dimensional (3D) images of thick tissue sections is an important basic capability necessary for many biological research studies. However, segmentation is often difficult because of the tight clustering of nuclei in many specimen types. We present a 3D segmentation approach that combines the recognition capabilities of the human visual system with the efficiency of automatic image analysis algorithms. The approach first uses automatic algorithms to separate the 3D image into regions of fluorescence-stained nuclei and unstained background. This includes a novel step, based on the Hough transform and an automatic focusing algorithm to estimate the size of nuclei. Then, using an interactive display, each nuclear region is shown to the analyst, who classifies it as either an individual nucleus, a cluster of multiple nuclei, partial nucleus or debris. Next, automatic image analysis based on morphological reconstruction and the watershed algorithm divides clusters into smaller objects, which are reclassified by the analyst. Once no more clusters remain, the analyst indicates which partial nuclei should be joined to form complete nuclei. The approach was assessed by calculating the fraction of correctly segmented nuclei for a variety of tissue types: Caenorhabditis elegans embryos (839 correct out of a total of 848), normal human skin (343/362), benign human breast tissue (492/525), a human breast cancer cell line grown as a xenograft in mice (425/479) and invasive human breast carcinoma (260/335). Furthermore, due to the analyst's involvement in the segmentation process, it is always known which nuclei in a population are correctly segmented and which not, assuming that the analyst's visual judgement is correct.     

摩托车链板的机器视觉检测

针对摩托车用滚子链外链板的几何尺寸及圆度误差提出了一种基于机器视觉的检测方法。用数字摄像头采集数字图像,通过IEEE1394数字接口卡将采集到的图像传输到计算机。对原始图像进行图像预处理及图像分割。在测量尺寸时,依据图像分割所提取的轮廓,使用自编程序利用最小区域法对圆度误差进行评定,并计算、测量链板的孔径和节距。在同一系统中实现了尺寸及形状误差的非接触测量,分析了测量误差的原因。分析及试验表明,用该方法对摩托车滚子链的几何尺寸及圆度误差进行测量及分析评定是可行的,并具有高效性和实用性。     

基于MATLAB数字图像边缘检测算子的研究

边缘检测是数字图像处理中的重要内容,是图像数据分析、模式识别、计算机视觉等技术的基础。因此对边缘检测算子的研究具有重要的意义。主要研究了基于MATLAB的数字图像边缘检测算子,其研究对象主要是灰度图像,通过对图像边缘检测算子阈值、方差等因素的仿真分析比较,获得各算子的适用范围,以便更好地灵活运用。