生物芯片图像样点的自动识别

样点自动识别是生物芯片信息自动提取的关键。根据样点、噪声和背景特征的关系提出一种新的自动识别方法。使用数学形态学和均值算子相结合的方法实现图像的滤波增强和背景亮度的估计;通过对功率谱的分析实现图像的倾斜校正和样点中心的网格定位;利用样点边缘亮度与均方差的特征实现样点中心和半径的校正。多幅生物芯片处理的实验证明该方法具有良好的抗噪声能力和弱信号辨识能力,能快速、准确地实现样点自动识别．     

基于图像的生物芯片点样质量检测方法研究

目的为了提高生产线上生物芯片点样质量检测的精度与效率,研究基于图像处理和卷积神经网络的算法,判断某生物芯片点样质量是否合格,并检测点样合格的生物芯片上的点样点半径。方法采用CCD相机获取生物芯片点样后的图像,通过图像预处理,利用canny边缘检测和圆的拟合等图像处理方法,得到点样点的几何信息,进而计算出点样点半径。同时提出基于卷积神经网络的点样质量检测方法,通过区域建议网络提取点样点卷积特征,引入分割全连接层来训练检测模型,通过离线训练来验证获得模型的最佳参数。结果和手动测量结果进行对比发现,半径误差不超过±0.1 mm,点样质量检测准确率为91.1%,单个生物芯片检测时间总和不超过1.6s。结论所提出的方法能够满足生产线上产品检测准确性和实时性的要求。     

基于行灰度投影相关的快速角度矢量估计算法

针对现存电子稳像算法估计图像旋转角度矢量的不足,提出了一种基于行灰度投影相关的快速图像旋转角度矢量估计算法。通过在当前帧图像中心位置选取灰度投影区域和参考帧图像相应区域作行灰度投影相关计算,估算出当前帧图像相对于参考帧图像的角度运动矢量。论述了灰度投影区域选取原则;分析了行、列灰度投影相关曲线的差异,得出行灰度投影相关曲线具有单峰性的结论;阐述了快速搜索算法的原理及其在本算法中的应用。实验表明,算法实现了对旋转图像序列角度运动矢量的快速估计（在PⅣ2.67GHZ,内存512M的PC机上,估算速度为37.7ms）,可对任意角度的角度矢量进行高精度的估计。     

悬浮式生物芯片检测中的图像处理研究

介绍了悬浮式生物芯片的特点和检测方法,首次提出了完整的悬浮式生物芯片图像处理解决方案。使用预处理去除图像中的噪声并对图像予以增强,然后通过模式匹配技术识别样点,确定位置坐标,定位误差小于1个像素。用数学形态学方法圈定样点范围,并用掩模技术提取信号强度值。使用整体或局部法处理背景。结合图形化开发环境LabVIEW和IMAQ工具包进行整个图像处理的集成和实现。结果表明,该方法能有效完成悬浮式生物芯片检测中的图像处理任务。     

溶胶-凝胶模板法制备BaTiO_3纳米点阵列及表征

通过结合溶胶-凝胶工艺和氧化铝模板制备了BaTiO3纳米点阵列.用XRD和原子力显微镜对其进行了成分和形貌分析.结果表明,通过改变基板温度,可以得到颗粒尺寸大小及分布在一定范围内变化的BaTiO3纳米点阵列,在基板温度为室温至80℃的条件下,制得的纳米点阵列颗粒尺寸为10～300nm,颗粒间距为70～900nm.与其他方法相比,此方法不需要昂贵的设备,且对基板没有特殊要求,还可以非常方便地用于其他多元氧化物纳米点阵列的制备.     

声散射预报中目标断面图像的快速旋转

Fourier衍射投影定理可用于水下目标重建和水声目标散射特性分析.将目标和周围介质的声学参数分布以二值图像描述,计算目标特性时需旋转目标图像而维持入射方向不变得到不同角度的投影.若直接用几何变换旋转二值图像,由于像素间的映射并非一一对应,会加重图像边缘的锯齿并在局部区域产生空值点.以极坐标系矩形网格下的平移实现目标旋...     

水下桥墩体内缺陷检测与定位的方法研究

为实现水下桥墩体内缺陷的检测与定位,提出一种将超声CT技术与质心定位技术相结合的方法。采用环形换能器阵列对桥墩截面以扇形束的扫描方式获取投影数据,使用滤波反投影算法重建截面的灰度图像,以实现内部缺陷的检测。鉴于声波在传播过程中会发生反射、折射等现象,利用区域生长算法分割出图像中的缺陷区域,进而求解出该区域灰度图的质心作为缺陷的位置点。计算机仿真结果表明,该方法能够重建较高质量的截面图像,并且能够快速识别体内缺陷所处位置。搭建检测系统对水下混凝土样块进行实验,较好地重建出样块中的缺陷,计算出缺陷定位的相对误差为4.6%,表明该方法切实可行并具有很好的实用性。     

CFRP感应加热线圈中心区域温度场

感应加热技术是实现碳纤维增强树脂复合材料(CFRP)低能耗高效固化成型的有效方法,提高CFRP感应加热温度场均匀性是保证成型质量的关键,而线圈中心区域温度场均匀性是保证材料整体温度均匀性的关键。根据电磁加热原理建立了CFRP有限元多场耦合的分析模型,通过对模拟计算和实验过程的温度场升温及分布情况的对比分析,证明了本仿真可以准确模拟CFRP感应加热温度场分布。根据图像的熵值理论将温度场均匀性通过熵值大小进行表示,实现了CFRP感应加热温度场均匀性的量化分析,并通过有限元模型计算研究了线圈直径及线圈与材料间距对线圈中心区域温度场均匀性的影响,得到了中心区域温度场均匀性与线圈直径及材料间距之间的关系曲线,为组合式线圈均匀加热CFRP提供了线圈直径及材料间距大小选择的理论依据。      

3D打印复杂点阵结构的缺陷三维可视化方法

针对3D打印的复杂点阵结构容易出现裂纹、未熔合或孔洞等缺陷,严重影响结构件的功能性能问题,开展了对3D打印的一种复杂点阵结构件的缺陷三维可视化检测方法研究。基于CT图像中结构件内部缺陷的灰度值差异特征,采用集合灰度值法自动识别一类缺陷并分割提取,由光线投射法对分割得到的缺陷序列图像进行三维重构。实验结果表明:所提方法有效获得了点阵结构件内部一种典型缺陷的三维可视图,从三维角度可对缺陷的形状、大小等形貌细节信息进行描述,为进一步分析缺陷对结构性能的影响提供了有力的依据。     

关于文字识别中两种倾斜校正算法的比较研究

本文针对字符识别中的倾斜校正（即调整字符角度）问题,提出了一种新的校正方法,先经过灰度化,二值化的预处理,再提取字符图像每列的边缘点,经过公示算法计算出这些点所在的参考直线的倾斜角度,这个角度就是字符图像的倾斜角度,本文将这种方法与常用的hough变换校正法进行了比较,实验结果表明本文的算法不仅校正效果更好,而且大量节约了系统资源和简化了计算难度。     

采用细胞神经网络快速校正生物芯片图像倾斜

倾斜校正是全自动生物芯片图像处理必不可少的环节。针对以往倾斜校正算法耗时过长,本文提出一种快速倾斜校正算法。算法首先用细胞神经网络（CNN）寻求各个样点的中心,然后进行Hough变换,再计算方差来寻求倾斜角,最后利用CNN灰度图像旋转模板进行倾斜校正。本算法利用了细胞神经网络并行处理的特性,并充分考虑了生物芯片图像的特点。理论分析与试验结果显示本文算法能够准确高速地完成倾斜校正。     

用傅里叶相移特性估计位移

提出了一种从频率域出发,估计出运动物体在空间域位移的算法。利用傅里叶变换的自配准性质和位移特性,用极坐标形式下连续图像相位谱的差直接估计运动目标的位移。与传统的寻找迪拉克峰值的方法相比,本方法无需重新变换回空间域,从而节省了处理时间,具有更好的实时性。实验证明,这种先求出相位谱的差,再用其周期数来估计位移的算法简单明了,其分辨能力不低于一个像素。该理论可以应用于图像跟踪和图像后期处理中。     

Application of wavelet modulus maxima in microarray spots recognition

This paper presents a novel approach to recognize the microarray image spots. The approach is based on the detection of wavelet modulus maxima in the microarray images. The detected maxima is actually the contour of the spots and thus the spots are recognized precisely. Then, the intensities within the contour of the spots can be obtained with low error rate. The test results on example image show this is an effective approach, especially for those spots with low intensities.     

Two-dimensional microarray of HepG2 spheroids using collagen/polyethylene glycol micropatterned chip

A new cell chip technology in the form of a two-dimensional microarray of HepG2 spheroids was developed by using the microcontact printing technique. The chip consisted of several collagen spots in a triangular arrangement within a 100-mm² area at the center of a glass plate (24 × 24 mm), which served as the cell adhesion area; the region excluding the collagen spots that was modified with polyethylene glycol (PEG) served as the non-adhesion area. HepG2 cells inoculated onto the chip gradually formed spheroids with smooth surfaces and high circularity on each collagen spot due to cell proliferation; the spheroid diameters remained constant after 10 days of culture. Such a two-dimensional microarray configuration of HepG2 spheroids could be maintained for at least 2 weeks. The spheroid diameter was directly proportional to the pitch between the collagen spots on the chip. This indicates that we can factitiously control the spheroid diameter. In addition, albumin secretion activity of HepG2 spheroids increased with the increase of spheroid diameter. This chip technology may be applicable as a cellular platform for developing two-dimensional spheroid microarrays.     

cDNA microarray image segmentation using root signals

A vector processing based framework suitable for cDNA microarray image segmentation is introduced and analyzed in this paper. By using nonlinear, generalized selection vector filters the framework proposed here classifies the cDNA image data as either microarray spots or image background. The solution converges to a root signal that represents the segmented cDNA microarray image with the regular spots ideally separated from the background and with their coloration uniquely described by dominant color vectors. It will be demonstrated that the framework readily unifies image denoising, enhancement, data normalization, irregular spot rejection, and spot segmentation in one processing step delivering excellent performance at reasonable computational cost. © 2006 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 16, 51–64, 2006     

一种精确的光点测试系统

介绍了一种精密光点测试系统。它采用脉冲方法刷新光点,并且刷新频率和电源频率保持同步,从而避免了扫描及视频方法引起的测量误差,同时避免了散逸电磁场的影响;并克服了采用扫描光栅、通过信号发生器或者测试软件产生测试信号来进行测试的系统存在的问题。该测试系统尤其适合电子枪和偏转线圈设计人员进行精密的光点测试分析。     

Laser Spot Welding of Thermoplastic and Ceramic: An Experimental Investigation

Laser welding of dissimilar materials classes has vast potential in a variety of applications due to the non-contact nature of the process. This study investigates the effects of CO₂ laser process parameters on the characteristics of spot welded joint between ceramic and thermoplastic. Using a full-factorial design an experimental investigation is carried out studying the effect of welding parameters (laser exposure, number of spots, stand-off distance) on the tensile strength and the diameter of spot weld. The study reveals an increase of joint tensile strength with the increase in the number of spots and the duration of exposure. However at low laser power, the effect of stand-off distance on the spot weld diameter is found to be inconclusive. Optimal process parameters are estimated using the multi-objective optimization on the basis of ratio analysis. The analysis shows that the higher number of spots and the longest duration of laser exposure result in optimal weld characteristics. Alongside these conditions, a larger stand-off distance is seen to have beneficial impact on the weld characteristics.     

Optical grating diffraction method: from strain microscope to strain gauge

The grating diffraction method for direct strain measurement is reviewed. Two systems which use this method are presented. The first system is a compact strain microscope. A Leitz optical transmitting microscope with white light source is reconstructed by developing a loading and recording system. Gratings with median density of between 40 and 200 lines/mm are used. With the help of a Bertrand lens, the Fourier spectrum of the grating is formed with high image quality on the CCD sensor plane. Software is developed to precisely, quickly and automatically determine the diffraction spot centroids. The second system is a new strain sensor based on a high-frequency grating and two Position Sensor Detectors (PSDs). The grating, attached to the surface of the specimen, is illuminated by a focused laser beam, generally with a frequency of 1,200 lines/mm. The centroids of diffracted beam spots from the grating are automatically determined using two PSD sensors connected to a personal computer. The shift of diffracted beam spots due to specimen deformation is detected. Strain sensitivity of one micro-strain can be obtained, as can a 0.4 mm spatial resolution for strain measurement. The system can be used for both static and dynamic tests.     

激光偏角测量技术研究

介绍一种采用激光测量反射镜偏角的方法 ,它适合于光学系统装配时的测量。采用 CCD细分技术对成象光斑质心进行提取 ,提高了偏角测量精度 ,在 1 .2 5°的测量范围内 ,可以达到± 3″的测量精度     

一种基于小波变换的目标融合跟踪方法

本文在分析小波变换的基础上，将小波分析应用到目标图像的融合跟踪技术上，利用小波的多尺度和多分辨特性，不仅能够获得不同分辨力下的图像序列，进行目标图像融合；还能有效地从信号中提取突变信号。对函数或信号进行多尺度的细化分析。图像边缘用小波变换进行处理和提取并对图像形心进行计算。能够得到较好的轮廓提取效果和形心定位精度，进而说明了小波变换可能成为目标跟踪中较好的数学方法。