结核杆菌涂片显微视觉检测系统的自动聚焦

结核杆菌医学涂片大多具有观察区内容稀疏不均匀、杂质较多的特点,使用自动显微镜检方法进行图像采集时,会出现清晰度区分困难、效率低下、甚至聚焦评价失效的问题,为提高自动镜检的效率和准确度,本文自主搭建了显微视觉计算机自动检测系统,对结核杆菌涂片的自动聚焦技术进行系统的研究。首先,对比研究11种常用聚焦函数对结核杆菌镜检玻片图像聚焦评价的优劣,并分析了聚焦成功和失效的原因。在综合分析各聚焦函数对结核杆菌涂片的聚焦效果基础上,提出了一种基于Tenengrad的改进型聚焦评价函数,通过改进内容像素的聚焦权重提高聚焦准确度,优化图像处理算法来提高图像采集效率。实验结果表明:改进型Tenengrad聚焦函数FTen-Q在结核杆菌涂片的各类视野图像评价方面具有高灵敏度和准确度,其聚焦成功率和运算效率分别提高了13.884%和17.616%,可以满足结核杆菌涂片类非均匀涂片的显微视觉自动检测应用要求。     

数字共焦显微仪序列光学切片自动采集方法研究

提出一种基于图像处理的数字共焦显微仪序列光学切片自动采集方法。方法分两个步骤:第一步是对细胞进行自动聚焦;第二步是细胞的起始点自动调焦定位及进行光学切片。方法采用清晰度评价函数作为判别标准,将灰度差分算法与拉普拉斯函数法相结合,通过计算机实现两个步骤的自适应调焦控制,进而实现光学切片的自动采集。实验结果表明,方法有较高的自动聚焦和细胞起始点定位的速度和准确性,对三维球形细胞序列光学切片图像采集,简单易行,效果良好。     

基于小波分析的显微序列图像合成系统

由于显微镜物镜焦深范围有限,故造成显微图像局部模糊,研究人员对图像的理解不完整等问题。根据聚焦评价函数分布,采用小波分析,将一序列局部聚焦图像合成全焦图像。改进的小波融合算法得到的融合图像极大地保留了系列图像聚焦部分的信息,合成全焦图像。由此得出的聚焦合成图像使研究人员可直接测量所需的显微结果。     

全自动涂片显微视觉扫描仪控制系统开发

面向医疗显微形态学检验需求,设计了一种通用型全自动涂片显微视觉扫描仪的机械平台及自动检验工艺流程,构建了"计算机+工业相机+嵌入式控制器"的全自动涂片显微视觉扫描仪控制系统;开发了基于ARM的嵌入式控制器,实现了多轴精密运动控制、多传感器反馈、二维码扫描、CAN通信等功能;基于高速相机和精密运动系统,开发了高质量图像自动聚焦高效采集算法;设计了控制系统上、下位机应用软件。将该系统用于结核杆菌和外周血涂片自动扫描检验实际应用中,结果表明:全自动涂片显微视觉扫描仪控制系统运行稳定,图像采集清晰高效,操作简单,可以满足多数医疗镜检项目的使用需求。     

超声检测三角矩阵聚焦成像算法

超声相控阵技术近些年发展迅速并被广泛应用于工业无损检测领域,线阵探头全矩阵聚焦方法具有成像信噪比好、缺陷分辨力高等优势,近些年获得广泛研究,但其仍然存在采集数据量大、计算效率低的问题,使其在实际工业应用中受到限制。为了解决上述问题,基于检测系统收发通道的互易性提出三角矩阵聚焦成像算法,该方法简化了全矩阵聚焦成像算法的数据采集和成像运算过程。为验证聚焦成像方法的有效性,分别对类裂纹、平底孔和横通孔缺陷试块进行数据采集成像试验和信噪比分析,数据采集试验结果表明收发通道互换前后两信号具有很好一致性。进一步的数据比较分析表明,在成像质量方面,三角矩阵聚焦成像算法与全矩阵聚焦成像算法两者的图像信噪比变化不大;在数据采集和成像效率方面,三角矩阵聚焦成像算法降低了近一倍数据量,有效提高了计算效率。     

一种快速简单多焦面显微图像融合方法

开发了一个用于多焦面显微图像融合的插件,该插件能支持任意大小的图像处理,由输入的多幅不同聚焦面的局部清晰显微图像,融合得到完整清晰的图像。重点研究了基于小波变换的多焦面显微图像融合的算法,通过Matlab仿真确定较优算法,采用C++语言编写插件,实现图像融合的功能。该融合插件操作简便,融合效果好,用软件的方法扩大了显微物镜的焦深。     

激光扫描共聚焦显微镜技术的发展及应用

激光扫描共聚焦显微术是先进的分子和细胞生物学研究技术。它在荧光显微镜成像的基础上加装激光扫描装置,结合数据化图像处理技术,采集组织和细胞内荧光标记图像。在亚细胞水平观察钙等离子水平的变化,并结合电生理等技术观察细胞生理活动与细胞形态及运动变化的相互关系。由于它的应用范围较广泛,已成为形态学、分子细胞生物学、神经科学和药理学等研究领域中很重要的研究技术。     

显微视觉系统的自动聚焦及控制

对自动显微镜的自动聚焦评价函数及聚焦控制策略进行了研究.首先,介绍了频域聚焦函数提升小波变换及时域聚焦函数Sobel-Tenengrad算子,通过将提升小波变换和Sobel-Tenengrad算子有机组合提出了一种新型聚焦评价函数.然后,利用离焦、正焦样本图像对自组织算法进行无监督训练,使用粒子群优化算法加速训练过程,并以经过学习的自组织映射算法作为聚焦控制器.最后,进行了显微视觉自动聚焦实验.实验结果表明:新型组合算子具有单峰性,峰值处变化陡峭,对不同样本、不同倍数物镜均可在正焦位置达到最大值,鲁棒性强;经过学习控制器后平均仅用7.6步即可完成自动聚焦,与爬山法相比,该聚焦算法不仅大大提高了聚焦速度且性能稳定,对每幅输入图像处理、识别时间约为120ms;满足了显微视觉自动聚焦要求,获得了良好聚焦效果.     

简单图像的快速聚焦

为了有效地实现简单图像的快速高精度自动聚焦,提出了一种新的快速聚焦算法。首先在相机进行调焦时自动获取系列零件图像,并记录各图像对应的位置;然后计算每张图像中目标边缘的灰度变化跨度值,从中找到跨度值最小的图像,即为最清晰图像;最后把其所在位置反馈给硬件驱动系统,实现自动聚焦。用新算法分别对不同形状和不同材料的零件,在添加椒盐噪声和没有添加噪声的情况下进行实验,并与计算量小的几种经典最优聚焦函数做了对比试验。结果表明,用新算法对简单图像进行聚焦比常用的最优聚焦函数更敏锐,单峰性更好,抗噪能力更强,而且速度比最快的绝对梯度函数快30%以上。因此,新算法在拍摄简单场景时,鲁棒性好,可以更好地实现快速聚焦。     

自动聚焦技术在刀具测量仪器中的应用

在光学测量仪器中,如何提高聚焦精度是非常重要的。本文针对刀具几何参数测量仪,设计了自动聚焦系统。提出了以步进电机驱动显微镜移动,结合图像处理技术来实现自动聚焦的方法。通过对比实验,选择了图像的灰度梯度向量模方和算子作为图像是否聚焦的评价函数,以及给出了登山式自动聚焦策略。最后对该系统做了重复性聚焦实验,证明上述方法可以提高系统的聚焦精度。     

PCA-based method for 3D shape recovery of microscopic objects from image focus using discrete cosine transform

This article introduces a new algorithm for shape from focus (SFF) based on discrete cosine transform (DCT) and principal component analysis (PCA). DCT is applied on a small 3D neighborhood for each pixel in the image volume. Instead of summing all focus values in a window, AC parts of DCT are collected and then PCA is applied to transform this data into eigenspace. The first feature, containing maximum variation is employed to compute the depth. DCT and PCA are computationally intensive; however, the reduced data elements and algorithm iterations have made the new approach competitive and efficient. The performance of the proposed approach is compared with other methods by conducting experiments using image sequences of a synthetic and two microscopic objects. The evaluation is gauged on the basis of unimodality, monotonicity, and resolution of the focus curve. Two other global statistical metrics, root mean square error (RMSE) and correlation have also been applied for synthetic image sequence. Besides, noise sensitivity and computational complexity are also compared with other algorithms. Experimental results demonstrate the effectiveness and the robustness of the new method.     

基于动态因子自适应搜索算法的自动调焦研究

为提高自动调焦的性能,采取动态因子自适应搜索算法。首先,建立分辨率逐步递减的调焦窗口,以减少非聚焦区域对调焦评价函数计算的影响;其次,通过动态调焦因子方法修正目标变化对清晰度评价函数的影响;最后,通过适应变步长爬山算法结合清晰度评价函数搜索聚焦点,提高搜索速度和调焦精度。试验仿真显示,本文提出的算法获得的图像清晰,其性能受噪声影响小,可满足自动调焦中对稳定性、实时性和清晰度的要求。     

一种变物距的自动对焦方法

针对像距无法改变的特定环境,提出了一种变物距的自动对焦方法。首先通过设置标志矩形,采用SUSAN算法中的角点获取算法,取得对焦窗口;其次建立对焦窗口转换函数,使不同物距下的对焦窗口包括相同的成像信息;最后使用能量梯度对焦评价函数,取得最佳对焦位置。实验表明,该系统适用于像距无法改变的环境,并且具有良好的自动对焦精度、速度以及鲁棒性。     

The development of an automated PIV image processing software—SmartPIV

Since the popularity of digital particle image velocimetry technique (DPIV), many PIV image processing algorithms have been proposed. Amongst them, fast Fourier transform (FFT) Cross Correlation, Discrete Window Offset Cross Correlation, Iterative Multigrid Cross Correlation, Iterative Image Deformation Cross Correlation and cross correlation based particle tracking methods are widely used algorithms and have been extensively studied by researchers. All of these algorithms have their advantages and disadvantages in terms of computational load and measurement accuracy. To choose a suitable algorithm, researchers not only need to understand the complex principles of these algorithms, but also need to find out their applicable flow conditions. This could greatly increase work load for PIV users who focus more on flow structure itself instead of PIV algorithms. It is therefore necessary to develop a method which can choose PIV algorithms wisely according to the input PIV images. This paper firstly reviews the development of PIV algorithm with mainly focus on analysing advantages and disadvantages of six widely used algorithms. By using both synthetic and real PIV images, comparative studies are then carried out among these algorithms. The tests give a rate for the performance of the algorithms and provide a parameter to automatically separate pattern match and particle tracking algorithms. Based on qualitative and quantitative analysis, an automated PIV image processing method—SmartPIV is proposed and tested by both synthetic and real PIV images. For all the three test cases, the SmartPIV successfully picks the most suitable algorithm and gives very promising results.     

含三次位相元件照相物镜的设计

设计了一个光学/数字成像系统,该成像系统利用三次位相元件辅以数字图像处理技术,在保持系统光通量及分辨率不变的前提下,提高了光学系统焦深。进行了仿真实验,实验以传统的双高斯型照相物镜为光学系统模型,选取焦点及离焦-1.6λ(λ=550 nm)近似-6.4倍焦深两个成像位置。对比分析了传统光学系统及光学/数字系统的焦点位置调制传递函数(MTF)、离焦1.6λ处调制传递函数(MTF)及在瑞利判据±100倍焦深范围内的离焦传递函数。通过光学设计软件CODEV9.7新增图像模拟功能,得到了光学系统的模拟成像结果。利用自编图像处理程序,得到最终的成像结果。结果表明,该光学/数字成像系统确实能够有效扩大光学系统焦深6倍以上。     

一种自动对焦算法的优化

随着数码相机的飞速发展,在实际应用中对数码相机的易用性和智能化要求越来越高。为了提高图像采集的准确性和简便性,开展了自动对焦技术和算法的研究。通过对对焦平台的搭建,在评价函数一定的情况下,对爬山法驱动寻址算法步长减半进行了优化,在保证准确性的情况下,更快地实现了自动对焦,减少对硬件的要求,节省资源。结果表明,该优化算法有一定的优越性。     

基于图像处理的微装配自动调焦系统

介绍了一个基于图像处理的自动调焦系统,该系统由光学显微镜、CCD、力矩电机和齿轮传动机构等组成。调焦系统中采用粗/精结合的调焦方法,即首先采用基于Krisch边缘检测算子的清晰度评价函数对被测物体进行粗调焦,并在计算机上采集了包含目标的大致轮廓和边缘的显微图像;然后针对全景图像上的某个目标区域采用基于高频分量的清晰度评价函数进行精调焦,使得显微图像显示出更多的纹理细节。实验中清晰度评价函数算法均采用Windows下的Visual C++实现。实验结果表明,该自动调焦的系统精度可达±4.8 μm,基本上满足了微装配任务的调焦精度要求。     

基于双拟合优化的聚焦深度三维形貌测量方法

聚焦深度法广泛地应用于小尺寸零构件的三维形貌测量。 针对聚焦评价曲线多存在噪声导致三维形貌测量的精度下 降,且效率受图像序列数目和聚焦评价算法限制的难题,通过结合双拟合优化理论与聚焦单峰性评价指标提出了一种基于自适 应权重的聚焦极值搜寻方法,并进一步基于三次多项式二次插值算法和多级中值混合滤波优化深度信息。 实验结果表明,方法 对仿真图像数据的均方根误差(RMSE)较高斯拟合和多项式拟合分别降低了 19. 08% 、17. 32% , 对球栅阵列封装 (BGA) 图像 中受噪声干扰较大的区域仍具三维分辨能力,同时对钻削刀具进行了测量,提出的方法在 50% 的图像序列数目和图像分辨下 RMSE 仍降低了 77. 8% 、62. 6% ,有效减少了聚焦曲线噪声对测量结果影响的同时提升了三维测量的效率。     

Water meter pointer reading recognition method based on target-key point detection

With the rapid development of video image technology and fifth-generation mobile network technology, the automatic verification of mechanical water meters has become an increasingly important topic in smart cities. Although much research has been done on this subject, the efficiency and accuracy of existing water meter pointer reading technology can still be improved. This paper proposes a new water meter pointer reading recognition method based on target-key point detection. Our method consists of a target detection module and a key point detection module. The target detection module uses a modified YOLOv4-Tiny network to detect and classify the areas where the dials and pointers are located in water meter images with distinct characteristics. The key point detection module is used to detect the key point of the pointer image. In this module, the structure of the RFB-Net network is improved to introduce multiple layers of low-level feature information, therefore, it can make full use of the information between multi-scale feature layers for key point detection. In addition to, aiming at the problem of dial rotation, a method of establishing a right-angle coordinate system based on key point is proposed to realize pointer reading. The whole method proposed in this paper is compared to the Hough transform feature matching algorithm and traditional machine learning algorithms through experiments which test the detection and recognition of the water meter dial, pointer and key points. The experiment results show that the missed detection rate of the model in this paper is 1.88% and 1.07% for the dial region and the pointer region, respectively. And the accuracy rate reaches 98.68%, the average processing time per image is 0.37 s. This implies that the water meter inspection task is completed quickly and accurately with strong robustness. Thanks to the lightweight algorithm of our approach, the model can also be fully automated and easily deployed on mobile devices.     

A novel autofocusing method using the angle of hilbert space for microscopy

Autofocusing technology is indispensable for routine use of microscopes on a large scale in biological field. The autofocusing method using the angle of Hilbert space is brought forward to measure whether the image is focused or not. The angle of Hillbert space can be used to evaluate accurately the similarity degree of two images. The experiment results show that the autofocusing method can decrease the computational cost and get accuracy for real‐time biological and biomedical images with noise robustness. The focus curves are smooth and possess the unimodality, the monotonicity and the symmetry. Compared with other classic and optimum focus method, the Hilbert method demonstrates its robustness to noise and can improve the focus speed. The experiments showed that the proposed method can increase the overall performance of an autofocus system and has strong applicability in various autofocusing algorithms. Microsc. Res. Tech. 77:289–295, 2014. © 2014 Wiley Periodicals, Inc.