权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Long Depth of a Focus Optical Microscopy Based on Computer Image Processing

基于计算机图像处理的长焦深光学显微镜

基本信息

批准号：
06452163
负责人：
MITSUI Kimiyuki
金额：
$ 4.74万
依托单位：
KEIO UNIVERSITY
依托单位国家：
日本
项目类别：
Grant-in-Aid for General Scientific Research (B)
财政年份：
1994
资助国家：
日本
起止时间：
1994 至 1995
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-06452163/
关键词：
Image processing Optical microscope Long focal length Hadamard transform CCD camera Surface roughness 焦点深度

项目摘要

There is a growing interest in micro electro mechanical systems. For the assembly operation of micro machining partsoptical microscope is used. It is often experienced that these operation under optical microscope is not easy, because depth of a focus of optical microscope is too short. From this point of view, a original method to get long depth of focus of optical microscope image using computer image processing has been proposed. The microscope objective is determined positioning by actuator such as PZT.At each located position, digital picture image through CCD camera is taken into a computer. The computer extractes has area where the image is in focus from these pictures, then it composes whole frame image. The depth of focus of the composite picture image is nearly equal to the sistance of motion of objective lens. As for the methods that distinguish between in-focus-area and out-focus-area, three different methods have been proposed. They are named here as differnce of luminance method, distributiuon of luminance method, and distribution of sequence of hadamard transfer method respectively. Processing time and accuracy point of view, distribution of sequence of hadamard transfer method is the best within these three methods.This method has been applied to the observation of several specimens such as standard surface roughness test-piece, grooved metal surface and so on, with excellent results. By adding displacement information to the recorded CCD camera image, this method is also applicable to the three dimensional surface topography measurement.

人们对微型机电系统的兴趣与日俱增。对于微加工零件的装配操作，使用光学显微镜。人们经常会发现，由于光学显微镜的焦深太小，在光学显微镜下进行这些操作并不容易。从这一角度出发，提出了一种利用计算机图像处理技术获取光学显微镜图像长焦深的独创性方法。显微镜物镜由PZT等致动器进行定位，在每个定位位置，通过CCD摄像机将数字图像送入计算机。计算机从这些图像中提取出图像焦点所在的区域，然后合成整个帧图像。合成图像的焦深几乎等于物镜的运动阻力。对于区分焦内和焦外的方法，已经提出了三种不同的方法。它们分别称为亮度法、亮度法分布和阿达玛变换法序列分布。从加工时间和精度来看，Hadamard传递法的序列分布是这三种方法中最好的，该方法已经应用于标准表面粗糙度试件、金属凹槽表面等多个试件的观察，取得了很好的效果。该方法通过在记录的CCD摄像机图像中添加位移信息，同样适用于三维表面形貌的测量。